ArticlePDF Available

Abstract and Figures

The overriding principle of ballot design is that it should not confer any a priori advantage to one candidate over another. Ballot format should not determine or condition an election outcome. Yet, there is a sizeable body of evidence which demonstrates that in many circumstances the design of ballot papers and voting machines contravenes the normative assumption of electoral neutrality. In this article, we look at the impact of ballot paper design at local elections in the Republic of Ireland (hereafter Ireland). The article uses data from an experimental election study conducted at the local elections in Ireland in 2009. Overall the study finds some evidence of a primacy effect and it also demonstrates that candidates located in the middle of the ballot face a challenge as they receive the lowest vote shares of all candidates across the four replica ballots. This mid-table obscurity remains even when party affiliation is known. Thus, it can be argued that candidates placed in such positions incur a disadvantage. To neutralise this effect, the article concludes with a recommendation that a system of random ordering of ballot positions across ballot papers should be implemented so as to ensure that each candidate appears at each ballot position on an equal number of times.
Content may be subject to copyright.
1
Ballot Paper Design, Evidence From an Experimental Study at the 2009 Local
Elections
FIONA BUCKLEY & THERESA REIDY
1
Department of Government, University College Cork, Western Rd, Cork.
ABSTRACT The overriding principle of ballot design is that it should not confer
any a priori advantage to one candidate over another. Ballot format should not
determine or condition an election outcome. Yet, there is a sizeable body of evidence
which demonstrates that in many circumstances the design of ballot papers and voting
machines contravenes the normative assumption of electoral neutrality. In this article,
we look at the impact of ballot paper design at local elections in the Republic of
Ireland (hereafter Ireland). The article uses data from an experimental election study
conducted at the local elections in Ireland in 2009. Overall the study finds some
evidence of a primacy effect and it also demonstrates that candidates located in the
middle of the ballot face a challenge as they receive the lowest vote shares of all
candidates across the four replica ballots. This mid-table obscurity remains even
when party affiliation is known. Thus, it can be argued that candidates placed in such
positions incur a disadvantage. To neutralise this effect, the article concludes with a
recommendation that a system of random ordering of ballot positions across ballot
papers should be implemented so as to ensure that each candidate appears at each
ballot position on an equal number of times.
Keywords: Ballot paper design, Ballot position, Irish elections, PR-STV
1. Introduction
The 2000 US Presidential election resurrected ballot paper design as a crucial aspect
of the mechanics of voting. Hanging chads and butterfly ballots entered mainstream
discussion of elections in a way that had not been expected. The overriding principle
of ballot design is that it should not confer any a priori advantage to one candidate
over another. Ballot format should not determine or condition an election outcome.
There must be a level playing field. Yet, there is strong evidence to suggest that in
many circumstances, the design of ballot papers and the operation of voting machines
contravenes the normative assumption of electoral neutrality.
1
Address for correspondence: Dr Theresa Reidy t.reidy@ucc.ie
2
The research discussed in this article underscores the fact that ballot position effects
exist and they are especially prominent under certain types of electoral systems and
types of ballot structure. This is not just a technical observation to be consigned to the
pages of election mechanics literature. Ballot paper effects raise important questions
about political representation. Some systems provide a strategic bias in favour of
particular candidates and this can result in distortions of representative democracy.
Edwards (2015) argued that US states which use alphabetical ballots select candidates
with surnames from the start of the alphabet in far greater numbers and he suggests
that this can have substantial outcomes for political representation. Kimball and Kropf
(2005) concluded that ballot design exacerbated racial disparity in US voting patterns
with some types far more likely to result in unrecorded votes. Similarly, Carman et al
(2008) found that the combination of poor ballot design and social deprivation led to
unusually high levels of spoilt votes at the 2007 Scottish Parliamentary elections.
Ballot design matters and it can have important consequences for the political system.
In this article, we look at the impact of ballot paper design at local elections in
Ireland. Election management is relatively moribund in Ireland but elections work
reasonably well and there is a high degree of trust in the process (Sinnott et al., 2008;
Buckley et al., 2015). However, resistance to change and suspicion of innovation
often greet attempts to modify electoral practices. This theme emerges in a number of
the contributions to this special issue (Coakley, 2015; Murphy, 2015; Farrell, 2015).
In Ireland, ballot paper design procedures are set out in electoral law and are common
across all candidate based elections; local, European, Dáil, Seanad and Presidential.
Candidates are listed on ballot papers in alphabetical order and are accompanied by
photographs of the candidates, party emblems, occupational information and
residential or professional address. Research from the late 1970s and 1980s indicated
that there were important advantages for candidates placed at the top of Irish ballot
papers, a phenomenon often referred to as the primacy effect. The conclusions were
clear but no action was taken at an administrative level to address this a priori
advantage. This article returns to the question of whether the design of ballot papers
has a conditioning effect on election outcomes. Coming nearly forty years after the
original work, it is timely to re-evaluate the extent of the primacy effect at Irish
elections. Ballot paper design has changed in the intervening period and further
3
questions have been raised about whether the inclusion of photographs and party
emblems have added new dynamics to the decision making of voters (Buckley et al.,
2007). Although these issues are outside the scope of this article, it is important to
note that the design of the ballot paper has evolved albeit with the alphabetical core
still intact.
The research uses data from an experimental election study conducted on the day of
the local and European Parliament elections in Ireland in 2009. The experiment
deployed sample ballot papers with actual candidates, and interviewed actual voters
from a different local electoral area at four polling stations. The research design is
unusual and while it presents some limitations on the extent of the statistical analysis
of the data, it is unusual to collect data in a manner so proximate to actual election
conditions and we see this as an important strength of the contribution.
The article is structured as follows; in section two we look to the international
research on ballot paper design to refine the hypotheses for the paper. Section three
provides a brief contextual overview of local elections in Ireland. Section four
outlines the details of the election experiment and the methods used in this paper. The
results are presented in section five. Section six discusses the results and recommends
the introduction of randomly order ballots to alleviate any potential for disadvantage
that candidates may incur under the alphabetical (ordinal) system of candidate listing.
2. Ballot Paper Design
Models of ballot paper design abound. Some, or all, of the following may appear on
the ballot; candidate names and personal information; party names, information and
logos; candidate photographs; party symbols or images; a single language or multiple
languages. Reynolds and Steenberger (2006) trace the evolution of the design of
ballot papers over the centuries and provide insights into the varied and unique
formats that can be found across the world. At its core, the work on ballot paper
design is interested in questions around how the structure, information, quality and
colour of ballot papers may influence voters in their decision making processes. An
important normative question dominates, do ballot paper designs deliver different
outcomes.
4
Reynolds and Steenberger (2006) provide a useful starting point for studies of ballot
paper design. They address four important questions; how are changes in the design of
the ballot processed within the political and administrative system; are ballot types
related to the type of political regime, can ballot design assist voters with literacy
problems and finally does ballot design affect the outcome of an election. Grounded
in a political psychology framework, their findings are particularly interesting. They
conclude that elaborate ballots are more common where literacy levels are low but
there is little evidence that the ballots assist voters with literacy problems. Indeed,
they demonstrate that levels of spoiled votes are higher in places with complex
ballots. In their final point, they conclude that ballots are a highly manipulative tool
which can influence election outcomes. While some of the elements of manipulation
they document might be more likely in emerging democracies, it is clear that ballot
design in established democracies also raises interesting and potentially serious
questions.
The work on ballot design in established democracies can be organised into a number
of strands. We provide an overview of the main points but the focus of our review is
on ballot position effects.
Candidate Photographs
Candidate photographs are placed on the ballot in many developing world countries.
They are generally introduced to ameliorate the effects of high levels of illiteracy and
as Reynolds and Steenburger (2006) have argued when combined with other
information on the ballot, candidate photographs can lead to a complex ballot.
Interestingly, candidate ballot paper photographs are used in a small number of
established democracies, Ireland and Portugal being examples of these. A central
objective of the work on ballot paper photographs has been to examine the ways in
which voters can use candidate photos as a type of shortcut to infer certain
characteristics or traits about the candidates (Rosenburg et al., 1986; Todorov et al.,
2005; Oliviola et al, 2012). Hermann and Shikano (2014) found that impressions of
attractiveness and competence influenced the political traits which participants
inferred from candidate photos and Johns and Shephard (2011) concluded that the
addition of photographs to the ballot for British elections could impact outcomes in
marginal constituencies (see also Shephard and Johns, 2008). With a focus on low
5
information elections, Banducci et al. (2008) concluded that voters include the
attractiveness of candidates as a factor in their decision-making. Ballot paper
photographs are used at Irish elections and they provided the conduit through which
this study was conducted.
Ballot Position
The central focus of this article is ballot paper positional effects. Generally, this work
confirms a tendency among voters to prefer candidates whose names appear at the top
of the ballot, compared with lower placed candidates. However, within that overall
picture, there are a number of factors which can influence the scale of the positional
effects. First we look at the general findings on positional effects before going on to
discuss the electoral context and the electoral system.
Koppell and Steen, in a study of the New York 1998 Democratic primaries, showed
that candidates received a greater proportion of the vote when listed first than when
listed in any other position. Strikingly, they found that in seven of those 71 contests
the advantage to first position exceeded the winner's margin of victory (Koppell and
Steen 2004: 267). Koppel and Steen treat voting as a cognitive task. “When voters are
faced with a choice among alternatives, they will conserve resources and select the
most accessible satisfactory option presented, even if it is not optimal…if choices are
presented visually, as in an election ballot, the first option presented is most
accessible and a "primacy effect" is expected…the magnitude of position bias
depends on how many voters do not have substantive bases for choice” (Koppell and
Steen 2004). This is consistent with psychological research on positional effects
which has shown that if a list of random words is flashed briefly on a screen, the first
is more often perceived correctly and more often recalled subsequently (Kelley and
McAllister 1984: 454).
The advantage of being placed in the first position on the ballot has been confirmed in
a series of papers (Faas and Schoen, 2006; Lutz, 2010; Meredith and Salant, 2013;
Marcinkiewicz and Stegmaier, 2015) although there is some variation in the extent of
the advantage. Lutz (2010) examined Swiss elections using an open PR ballot and his
results demonstrated that the ballot effect was quite strong and eclipsed incumbency.
He demonstrated that being at the top of the list was most significant and that by
6
position three, the effect had fallen off quite a bit. The primacy effect is also
confirmed by Meredith and Salant (2013) and they demonstrated that candidates listed
first on the ballot win elections between four and five per cent more often than those
placed in other positions on the ballot, all other things being equal. They also added
that the primacy effect is more pronounced in contests with more candidates on the
ballot paper. Looking at California, Ho and Imai (2008) reported that in primaries
major party candidates gained three points while minor party candidates could
actually double their vote share however, they found that at general elections only
minor parties benefited from being placed first on the ballot.
The type of electoral system in operation is an important consideration in studying
positional effects. Pilet et al. (2012) discussed the impact that ballot structure can
have on the constituency workload of members of parliament. However, Faas and
Schoen (2006) pointed out that candidate position is pre-determined through party
label in many systems so there is little that candidates can do to respond to the
primacy advantage. However, in preferential voting systems where voters may rank
their preferred candidate, 1,2,3, it should be expected that positional voting bias is a
feature but furthermore, parties and candidates might be expected to respond to the
incentives provided.
The ballot paper at Irish elections is structured by candidate alphabetical order. This is
important as it avoids some of the endogeneity problems identified by Lutz (2010)
and found in studies using list systems where pre-determined assessments of a
candidates electoral success play an important part on the decisions made by parties
on where candidates are ranked on the ballot. PR-STV offers a pure example of
alphabetical order where voters are in control. In their study of the 1973 Irish general
election, Robson and Walsh (1974: 191-203) found that candidates placed higher on
the ballot enjoyed a "distinct advantage" over their fellow candidates. They analysed
the number of votes gained by candidates of the same parties according to their
position on the ballot, and found that candidates placed at the top of the ballot
received more votes than their party colleagues. Their study also showed that position
effects were more prominent among non-incumbents than among incumbents.
7
Subsequent research has confirmed the findings of Robson and Walsh. Marsh (1987:
65-76) examined the impact of ballot position on election outcomes in the seven Irish
elections held between 1948 and 1982 and concluded that ballot structure was
important - candidates placed higher on the ballot had a clear advantage over their co-
partisans listed in subsequent positions. Bowler and Farrell (1991) presented
evidence of positional effects in the 1989 general and European elections
demonstrating that when given a choice, voters seemed to prefer to vote a party ticket.
They found that the amount of the preference schedule given over to preferences in
favour of Fianna Fail is much higher for those who place a Fianna Fail candidate first
than for voters in general. They demonstrated that this was also true for Fine Gael
voters (Bowler and Farrell, 1991).
The political context is an important consideration and provides the framework within
which voters make their decision. Following the adoption of PR-STV in Scotland for
the Scottish local elections, Curtice and Marsh (2014) compared the operation of the
PR-STV ballot structure at elections in Scotland and Ireland. They found strong
evidence of alphabetical bias in the Scottish data but in contrast to other studies of
Ireland, they reported weak positional effects from the Irish data. Still, the balance of
academic research is persuasive. There are strong indications that ballot position has
an impact. It follows directly then that candidates and parties might be likely to take
advantage of these effects.
Recognising the impact of name order in election outcomes, Ortega Villodres and
Garcia de la Puerta (2004: 3-14) highlighted consequences for the working of the
political system ‘it can affect party nomination strategies, and the conduct of electoral
campaigns; and it can be important in influencing the composition and behaviour of
deputies in Parliament’. Discussing the Australian case, Kelley and McAllister noted
that in the 1960s, the Democratic Labour Party was well known for regularly
nominating candidates with names at the start of the alphabet and they argued that
there is reason to believe that some of the other parties were also engaged in this tactic
(Kelley and McAllister, 1984). Spanish parties have also been known to embrace the
benefits of name order by manipulating the ballot papers for Senate elections “to
favour incumbents by placing their names first on the list” (Pereira and Villodres,
2002: 246). Hamilton and Ladd (1996) went further and suggested that Republicans
8
on North Carolina election boards in 1992 strategically chose the ballot form
expecting it to deliver positional advantages to their candidates.
It is entirely logical that political parties and candidates will alter their direct
behaviours in response to the clear evidence of primacy effects. Irish election lore is
littered with examples of candidates changing their names to get a position higher up
the ballot. Beverly Cooper Flynn (Mayo TD 1997-2011) is a recent example. She
opted for a double barrelled name upon marriage but unusually decided to put her
own surname last as her husband’s surname placed her on a higher point on the ballot.
Nicknames have been incorporated into family names such as in the case of Pat the
Cope Gallagher and Sean Dublin Bay RockallLoftus. Loftus was a Dublin based
councillor who changed his name to highlight political causes but the change had the
added advantage of raising his position on the ballot paper. Changing surnames from
English to Irish language versions and vice versa for ballot position advantage is also
present in popular memory of Irish politics.
Tying the strands of the literature together, the first hypothesis that we aim to test is
candidates placed at the top of the ballot will receive more first preference votes than
those placed those lower down on the ballot (H1).
Candidate Information
McDermott (2005: 201) suggested voters use shortcuts when making their decisions
on whom they will support in an election. She argued that voters economise, using
political and social stereotypes to judge candidates. Voters can use basic information
about candidates available on the ballot paper (or in election literature) - party
affiliation, incumbent/challenger status -. She goes on to point out that the voter "can
associate a candidate with a political and/or social group and project onto the
candidate such things as issue positions they believe the group holds" (McDermott,
1998:898). Popkin (1991) and McDermott (1998) examined the importance of
demographic cues on voter decision making in low information contexts and found
that candidate demographic cues are readily available to voters. McDermott noted
that a name on a ballot paper can indicate gender while a picture can inform a voter of
a candidate's gender, race, age and physical attractiveness. Using this information,
voters "are provided with stereotypical information that can help them choose
9
between candidates" (McDermott, 1998: 912) and McDermott concluded that voters
did use the available candidate information to make electoral decisions.
Incumbency is one of the more critical pieces of information which can be provided
on a ballot paper. In a number of jurisdictions (Switzerland, Ireland) candidate
occupation is listed on the ballot paper. In the Irish case, incumbents can list their
occupation as public representative giving a clear indication of incumbency.
Incumbency is a factor to which the literature on elections pays a considerable amount
of attention. The evidence is mixed. Incumbency remains an advantage in the Irish
political context (Benoit and Marsh 2008). In Ireland turnover rates for politicians are
low by international comparisons with more than 80 per cent of incumbents returned
but in Britain, Norris et al. (1992) suggested that the electoral boost is so small as to
be significant in only the most marginal of contests. Indeed, Murray (2005) argued
that far from providing an electoral boost, incumbents may actually be an electoral
liability, especially if they are associated with an unpopular government. Thus, the
literature on incumbency is somewhat mixed indicating that the effect of the ballot
paper information might vary and could be dependent upon the political context.
Party information is a vital cue for voters and it is probably the least controversial
piece of information placed on the ballot paper. In many systems, voters ballot
choice is entirely restricted to a party ticket. When choice is available, there are many
variables which can influence a voter’s decision and Campbell and Miller (1957) were
among the earliest who demonstrated that the type of ballot influenced the extent of
split ticket voting by voters. Several studies point towards increased participation
when voters have political party cues on the ballot (Bonneau and Loepp, 2014).
Partisanship is low in Ireland by international standards but it is still an important
feature to consider (Marsh, 2007).
Due to the design of the experiment used in this study, it is not possible to consider
incumbency effects but the role of partisanship is included and is the basis of the
second hypothesis which proposes that party affiliation moderates primacy effects
(H2).
10
Election Context - Low information elections
We can predict that the effect of ballot design will vary across elections as we know
from the voting behaviour literature that electoral context matters a great deal for
voters. Looking to the literature on second order effects, we can speculate that
positional effects should be more likely when other cues have less resonance. A
number of studies have focused on this dimension. Taebel (1975) argued that ballot
placement of candidates is an important structural feature in accounting for voting
patterns and he concluded that ballot position is especially critical in election contests
in which the candidates are relatively unknown. Miller and Krosnick (1998: 291-330)
showed that position effects are prominent in low information elections where party
affiliations are not listed, when races have been minimally publicised, and when no
incumbent is involved. Voters are more likely to select candidates placed higher on
the ballot. Low information dynamics will be exacerbated when voters are casting
ballots for several contests together and perhaps also dealing with initiative
propositions or referendum questions, scenarios which are common in both the US
and a number of European jurisdictions. This leads us to our third hypothesis which
proposes that the primacy effect is stronger in a low information election context
(H3).
3 Local Elections in Ireland
Local elections in Ireland are conducted using PR-STV. Voters rank candidates in
order of their choice, 1. 2. 3. PR-STV confers a high degree of choice on voters, they
are able to choose across, and within, parties and, among non-party affiliated
candidates. Since 1999 local elections are co-scheduled with elections to the European
Parliament. This decision was taken in an attempt to improve participation rates at
both these contests. That being said, they are clearly second order elections in the
sense of Reif and Schmitt (1980) and turnout tends to be lower than at national
elections. Average turnout at local elections between 1967 and 2009 is 58 per cent.
The 2009 figure was just marginally below this average coming in at 57.7 per cent.
The variation over the decades is evident from figure one.
11
Figure 1: Turnout at Local Elections in Ireland
0
10
20
30
40
50
60
70
1967 1974 1979 1985 1991 1999 2004 2009 2014
Local democracy in Ireland is particularly weak and it is often argued that it would be
better termed local administration. There is an imbalance in the system where elected
officials have few real powers and tend to be subordinate to the senior appointed
officials, especially the city or county manager renamed chief executive officers under
the provisions of the 2012 local government reform document Putting People First.
Localism and brokerage are central features of the political system and when
combined with the limited powers of councillors, it delivers local elections which tend
to be dominated by discussion of national political issues although voting frequently
displays heavy local characteristics.
The design of the ballot paper at local elections follows the same regulations which
govern other political contests. Photographs of candidates have been placed on ballot
papers since 1999. The decision to include photographs was informed by arguments
that voters with literacy difficulties would be assisted in their voting. Specifically,
photographs were also identified as a measure to alleviate a problem, specific to the
Irish context, of many candidates of the same name appearing on the ballot paper.
Research undertaken after the 1999 election confirmed a positive reaction of all voters
to the photographs and specific support for the measure from voters with literacy
difficulties (Lansdowne Market Research 2000). However, this research was based on
the assumption that voters recognise their politicians or local political candidates. The
12
research did caution policy makers that the measure could strengthen a candidate
centred bias in Irish elections. In response to this advice, a decision was taken in the
Electoral Amendment Act (2000) to include party logos to offset any increase in
candidate centred politics (Dáil Debates (21/2/2001). Research into voter recognition
of politicians has undermined the assumption of widespread recognition. A survey,
using photographs of members of the European Parliament (MEPs), undertaken after
the 1999 European elections found that less than half of the electorate recognised
three or more candidates after the election. The authors concluded that only a small
number of voters were equipped with sufficient information for the photographs to be
of assistance (Lansdowne Market Research, 2000). Almost identical findings were
presented from the study wave taken after the 2004 elections (Lansdowne Market
Research, 2005).
The Joint Oireachtas Committee on the Constitution undertook a review of the
electoral system in 2009 and it recommended that the ballot paper should be
examined especially in light of potential problems with positional bias and also
problems with the photographs. Difficulties with the party emblems have also now
been added to this list. Ireland has a large number of non-party candidates and it was
noticed in 2014 that voters were using the blank space where party emblems appeared
for some candidates to mark in their preferences (Dáil Debates, 2015). Despite the
recommendations from the 2009 review and more recent concerns, no review has
taken place and indeed, no review is planned.
4 2009 European and Local Election Experiment
The June 2009 European Parliament and local elections in Ireland were somewhat
unusual. The elections were politically charged as the financial crisis that befell the
country in September 2008 had taken hold and the government of the day found itself
facing unprecedented economic difficulties. While European Parliament and local
elections are frequently dominated by national issues, in the 2009 election, economic
affairs dominated to the exclusion of all else.
The research presented here is taken from a unique experimental design. Four replica
ballot papers were developed and deployed at polling stations in Cork city and county
13
on the day of the election in 2009. The replica ballot papers were developed using
pictures of actual candidates from a constituency in another part of the country.
Candidate names and personal details were removed from all ballots. The structural
and information details of the ballots are as follows:
Ballot 1 - Photos in ballot order
Ballot 2 - Photos and party logo in ballot order
Ballot 3 - Photos in random order
Ballot 4 - Photos and party logo in random order.
As noted the candidates were from a different electoral area than that surveyed and
candidates would have been unknown to the survey respondents. National political
representatives are precluded from holding local electoral office simultaneously in
Ireland under what was known as the abolition of the dual mandate. As a result there
are few nationally recognisable figures in local politics. One of the conditions for
receiving access to the ballot paper photographs was anonymity for the election
candidates. Candidate photographs are classified as personal data and each candidate
had to agree to release their image for the purpose of the study.
2
In all, there were
nine candidates on the ballot, five men and four women.
The survey was administered at four polling stations on the day of the election. Two
urban polling stations and two mixed rural polling stations were used. Survey
respondents were asked to give their own age, gender and citizenship. They were
then asked to ‘vote’ for the candidates on the replica ballot papers, rank ordering the
candidates in the same manner as they would under regular PR-STV voting
conditions. Finally, survey respondents were asked to outline the key factor that
influenced their first preference vote choice.
In total a sample of 1201 ballots was achieved. The total registered electorate at the
four polling stations was 8342, resulting in a sample size of 14.39%. The refusal rate
was just over 3%.
2
Further information on the study is available from the authors.
14
There are a number of remarks that must be made in relation to the data collected.
First, and most significant, the respondents in the ballot experiment were actual
voters. Respondents were approached outside the polling station, after they had
completed their voting. Second the candidates on the sample ballot papers were actual
local election candidates. All of the data was collected on the same day. Furthermore,
in using actual voters and candidates, it was possible to compare the results of the real
election with those of the survey. The use of actual candidates and voters enhances
the external validity of the research. Tables one and two provide some summary
information on the candidates in the study under the two different formats used.
Table 1 Candidate Information Ballot (Alphabetical) Order
Candidate Label
Gender
Party Affiliation
Position 1
Male
Fine Gael
Position 2
Male
Fianna Fáil
Position 3
Female
Green Party
Position 4
Female
Fianna Fail
Position 5
Female
Independent
Position 6
Female
Sinn Féin
Position 7
Male
Independent
Position 8
Male
Fine Gael
Position 9
Male
Labour
Table 2 Candidate Information Random Order
Candidate Label
Gender
Party Affiliation
Position 1
Male
Fine Gael
Position 2
Male
Fianna Fáil
Position 3
Female
Green Party
Position 4
Female
Fianna Fáil
Position 5
Female
Green Party
Position 6
Female
Sinn Féin
Position 7
Male
Independent
Position 8
Male
Fine Gael
Position 9
Male
Labour Party
There are three hypotheses for the research.
Hypothesis 1: Candidates placed at the top of the ballot will receive more first
preference votes than those placed lower down on the ballot.
Hypothesis 2: Party affiliation will moderate primacy effects.
Hypothesis 3: The primacy effect is stronger in a low information election context.
15
In section five we present the results from our analysis. We look at the candidate in
the first position and also at the vote share outcomes for candidates placed at the end
of the ballot. We present the first preference vote share achieved by each candidate
and we also compare the results to the first preference share of the votes received by
the candidates from the real election in which the candidates participated.
5 Ballot Position Effects
We begin by looking at hypothesis one, candidates placed at the top of the ballot will
receive a higher percentage of first preference votes. Figure two presents the share of
the first preference vote received by each candidate on each of the four sample ballot
types. There are interesting impressions from the distribution of the votes. The pattern
is very mixed and while there is some evidence of a primacy effect for the candidate
in position one, the candidate in position three also does quite well, indeed as do the
candidates in positions eight and nine. The mean vote share for the candidate in
position one is 13 per cent but this figure is exceeded for candidates in positions three,
eight and nine. The highest mean vote share is for position nine at nineteen per cent
and this includes a range of 14 to 32 per cent.
Figure 2: Positional Effects % of the First Preference Vote (4 Ballot types)
We now turn to what we term mid-table obscurity. The mean share of the first
preference vote in position five is six per cent, position six is eight per cent and
16
position seven is seven per cent. The three positions have the lowest mean vote
shares. This pattern is most pronounced in sample ballot two (figure three).
Figure 3: Sample Ballot Two (Alphabetic Order: Photo and Party)
0
5
10
15
20
25
Ballot Two (alphabetical, photo,
party) % of 1st Pref
Another way of looking at mid table obscurity is to examine the ballot positions
which were most likely to receive no preference at all. This data is presented in table
three which reports frequencies. The ballot types are collapsed into two categories to
aid presentation of the information (alphabetical order and random order). Again,
there is a tendency for candidates placed in positions four, five and six to not have
received any preference at all, although it also evident that position two delivers a
very high level of no preference.
Table 3 Frequencies - Preference 1 and No Preference
Alphabetical Order
Random Order
Number 1
Preference
No
Preference
at all
Number 1
Preference
No
Preference
at all
Position 1
125
209
38
205
Position 2
71
297
38
233
Position 3
115
237
64
148
Position 4
65
258
47
211
Position 5
32
306
32
209
Position 6
39
297
53
191
Position 7
46
276
35
204
Position 8
97
210
69
176
Position 9
104
209
112
145
17
The data in table three also confirm the overall effect that the first and last positions
obtain very high shares of the first preference vote. Adding some nuance to the
literature, the evidence from the experimental study is that there is a bonus for
candidates placed both at the top and at the end of the ballot. Consequently,
hypothesis one is accepted but with some qualification that there are advantages for
candidates both at the top and at the end of the ballot.
Turning next to hypothesis two, we compare how political party information on the
ballot alters the overall share of the first preference vote. From figure four, it is clear
that there are significant differences in the vote share for candidates in positions 1, 3
and 8 across the two ballots arranged in alphabetical order. Candidates 1 and 8 were
both from Fine Gael and they saw their vote share increase sharply when this was
known to voters. This is consistent with the overall performance of the party at the
2009 local elections. The Fianna Fáil candidate in position three saw a sharp drop in
her vote once party was known again reflecting wider party performance in the
election.
Figure 4: Ballot Position Effects Alphabetical structure
The pattern in the sample ballots which used a random ballot structure also reflect the
wider party performances in a slightly different way but are still consistent with the
overall performance of parties at the real election. Both Fianna Fáil candidates
18
(positions 2 and 4) saw their vote share reduce with candidate four experiencing a
particularly sharp reduction. Both Fine Gael candidates also saw a small drop in their
vote shares but the Labour Party candidate’s vote share doubled as soon as his party
label was known (see figure five).
Figure 5: Ballot Position Effects Random structure
In an attempt to explore the question of primacy effects and party affiliation a little
more, we present data in figures six and seven on the parties which had more than one
candidate in the race. This allows us to explore the effect of ballot position within
party classifications. While there is a clear ballot position effect for the first Fine Gael
candidate in the ballot arranged in alphabetical order when party is known (figure 6),
the opposite occurs when the data from the random ballot structure is examined. At
this point it must be mentioned that there is a significant age difference between the
two candidates and it may be that the issues raised in relation to decisions on
candidate image may need to be considered. Unfortunately, this type of analysis is
outside the scope of this research.
19
Figure 6: Alphabetical Structure, Party Effects
Figure 7: Random Structure, Party Effects
As the data is drawn from four separate experiments, there are limits on the extent of
the statistical examination which can be undertaken. However, we can compare
means for each ballot position across the four ballot types and this is done in tables
four (a) and (b) for position one and in tables five (a) and (b) for position nine.
20
Table 4 (a) Comparing Means Candidate A
Ballot One
(Alphabetical,
Photo)
Ballot Two
(Alphabetical,
Photo, Party)
Means
276.97
334.75
F Statistic
2.723
Sig
0.99
Table 4 (b) Comparing Means Candidate A
Ballot Three
(Random,
Photo)
Ballot Four
(Random,
Photo, Party)
Means
361.42
490.66
F Statistic
8.515
Sig
(0.004)*
A higher mean is reported when party information is included for both formats. This
confirms that party has an effect and, when it is known to voters the primacy effect is
moderated. In other words, the candidate in position one is more likely to be allocated
a greater number of (lower) preferences ie; 6,7,8,9. This effect is statistically
significant for the random order ballot. We also turn to the other end of the ballot and
the analysis is repeated for candidate I. Here again, the means are higher under both
ballot orderings indicating that party does have an impact.
Table 5 (a) Comparing Means Candidate I
Ballot One
(Alphabetical,
Photo)
Ballot Two
(Alphabetical,
Photo, Party)
Means
243.25
384.47
F Statistic
16.579
Sig
(0.000)*
Table 5 (b) Comparing Means Candidate I
Ballot Three
(Random,
Photo)
Ballot Four
(Random,
Photo, Party)
Means
277.47
324.7
F Statistic
1.302
Sig
0.254
This analysis presented here leads to the general conclusion that party label does
moderate ballot position effects. When the party label is known, the overall
21
performance of candidates altered to more closely reflect the outcome of the real
election.
The final hypothesis we look at is an exploration of how the primacy effect should be
stronger in a low information election. Here we present two figures, both of which
include data from the actual elections. Data limitations mean that only a very
descriptive discussion is possible.
Figure 8: Alphabetical Structure and real Election Result
Figure 9: Random Structure and real Election Result
22
The sample ballots with photos only provide the closest condition we can get to a no
information election for voters while we suggest that the ballots with party affiliation
provide a low information context, in that candidate information has been stripped
out. In Figure 8, a noteworthy point is that the survey respondents were considerably
closer to the real voters when party affiliation was known. In Figure 9, the same is
true for the candidate in position two but to a lesser extent for the other positions.
Unsurprisingly we conclude that context matters - the more information that is
available to the survey respondents the more closely the survey results reflect the
outcome of the election.
6 Discussion
The findings presented here suggest that primacy effects at elections in Ireland may
have moderated quite a bit since the early research in the 1970s. Consistent with
Curtice and Marsh (2014) we find moderate evidence of a primacy effect across the
four experimental ballot formats. Candidates in the first position do well but they do
not outperform those in other positions consistently across formats. There is also an
advantage for candidates located on the last position on the ballot. There is some
evidence that candidates located in the middle of the ballot face a challenge as they
received the lowest vote shares of all candidates on the four ballot variations. When
party affiliation is introduced, ballot position effects are reduced but there is still some
evidence of a mid-ballot obscurity.
The electoral context is a vital factor and the nature of the economic crisis in Ireland
in 2009 meant that parties mattered a great deal. The results for the ballots with
political party affiliations replicated the disastrous performance of Fianna Fáil and the
surge in support for Fine Gael and Labour, both of whom were in opposition at the
time. The more information that was presented to survey respondents, the closer they
came to replicating the results of the real election.
It was noted in section one that the overriding principle of ballot design is that it
should not confer any a priori advantage to one candidate over another. As an
extension of that it should not confer any disadvantage on candidates. Ballot format
should not determine or condition an election outcome. Ballot papers should be a
level playing pitch for all candidates. The phenomenon of mid-table obscurity
23
observed in this study presents evidence that candidates placed in such positions incur
a disadvantage. To neutralise this effect, the introduction of a set of randomly ordered
ballot papers should be considered. This matter has been raised in a variety of fora in
recent years notably in the review of PR-STV by the Joint Oireachtas Committee on
the Constitution in 2010 and again at the Constitutional Convention in 2013 when
participating citizens were asked to vote on ‘changing the alphabetical order of
candidates on the ballot paper’. On that occasion 67 per cent of the convention
members were in support of such a change. To date the Irish government has yet to
act on these recommendations (Buckley et al, 2015).
The operationalization of randomised ballots would require the development of a set
of ballots in each constituency along the lines of the so-called ‘Robson Rotation’.
The ‘Robson Rotation’ advises that the versions of the ballot paper produced is equal
to the number of candidates running in a constituency to ensure that each candidate’s
name appears in each ballot position an equal number of times. So for a constituency
of five candidates, five ballots would be produced along the following lines
3
:-
Rotation #1
Rotation #2
Rotation #3
Rotation #4
Rotation #5
Candidate A
Candidate B
Candidate C
Candidate D
Candidate E
Candidate B
Candidate C
Candidate D
Candidate E
Candidate A
Candidate C
Candidate D
Candidate E
Candidate A
Candidate B
Candidate D
Candidate E
Candidate A
Candidate B
Candidate C
Candidate E
Candidate A
Candidate B
Candidate C
Candidate D
However, the introduction of randomised and rotational ballots would pose significant
administrative challenges. Under Ireland’s current system of electoral management
where responsibilities are spread across a variety of agencies and arms of government,
such a major change to the structure and deployment of ballot papers would likely
meet with strong resistance by the various stakeholders involved. The introduction of
such a system of ballot papers would require detailed research and a pilot study,
resources (both financial and personnel), voter awareness campaigns and a dedicated
3
This is only an illustrative example. As noted by Hawkey (2008) positions on the
first rotation are drawn by lot).
24
oversight and monitoring body to ensure proper distribution and usage of ballot
papers. In the continuing absence of an electoral commission in Ireland the current
system of ballot papers is likely to remain.
Acknowledgements
We would like to thank the team of students from the BA Politics, BSc Government
and the PhD in Government at University College Cork who assisted us with the
collection of the data for this study. We are also grateful to Prof Neil Collins
(Nazarbayev University) for his collaboration with us in the design of the experiments
and in other avenues of investigation which we have pursued using this data. We
would also like to thank Prof Alex Todorov (Princeton University) for advice he
provided to us in relation to experimental design. Finally, we would like to
acknowledge funding received from the Royal Irish Academy under their Charlemont
Grants (formerly Mobility Grants) scheme.
25
References
Banducci, S., Thrasher, M., Rallings, C. & Karp, J.A. (2008) Ballot Paper
Photographs as Cues in Low-Information Elections, Political Psychology, 29 (6), pp.
903 917.
Benoit, K. & Marsh, M. (2008) The Campaign Value of Incumbency: A New
Solution to the Puzzle of Less Effective Incumbent Spending, American Journal of
Political Science 52 (4), pp. 874-890.
Bonneau, C.W. & Loepp, E. (2014) Getting things straight: The effects of ballot
design and electoral structure on voter participation, Electoral Studies, 34(1), pp. 119-
130.
Bowler, S. & Farrell, D.M. (1991) Voter Behaviour under STV-PR: Solving the
Puzzle of the Irish Party System, Political Behaviour, 13, pp. 303-320
Buckley, F., Collins, N. & Reidy, T. (2007) Ballot Paper Photographs and Low-
Information Elections in Ireland, Politics, 27 (3), 174 181
Buckley, F., Farrell, D., Reidy, T. & Suiter, J. (2015) Electoral Commission
Consultation Commission. Dáil Debates.
Carman, C., Mitchell, J. & Johns, R. (2008) The unfortunate natural experiment in
ballot design: The Scottish Parliamentary elections of 2007, Electoral Studies, 27(3):
pp. 442-459.
Campbell, A. & Miller, W.E. (1957) The Motivational Basis of Straight and Split
Ticket Voting, American Political Science Review 51, pp. 293-312.
Considine, J. & Reidy, T. (2015) Baby-steps, the expanding financial base of local
government finance in Ireland, Administration, 63 (2), pp. 119-145.
Curtice, J. & Marsh, M. (2014) Confused or Competent? How voters use the STV
ballot paper, Electoral Studies, 34 (1), pp. 146-158.
Dáil Debates (2001) http://debates.oireachtas.ie/dail/2001/02/index.asp (debate on 21
February 2001).
Dáil Debates (2015) Oral Questions to Environment Electoral Reform.
http://www.oireachtas.ie/documents/op/Apr15/Questions/pq220415.pdf
Department of the Environment, Heritage and Local Government, (2004) Circular
F.2/04, Dublin: Department of the Environment, Heritage and Local Government.
Department of the Environment, Community and Local Government, (2012) Putting
People First, Action Programme for Effective Local Government. (Dublin:
Government Publication Service).
Edwards B.C. (2015) Alphabetically Ordered Ballots and the Composition of
American Legislatures, State Politics and Policy Quarterly, (published online
http://spa.sagepub.com/content/early/2015/03/05/1532440015573265.abstract and
accessed May 2015).
Faas, T. & Schoen, H. (2005) The importance of being first: Effects of candidates’ list
positions in the 2003 Bavarian state election, Electoral Studies, 25 (1): pp. 91-102.
Hamilton, J.T. & Ladd, H.F. (1996) Biased Ballots? The impact of ballot structure on
North Carolina elections in 1992, Public Choice, 87, pp. 259-280.
Hawkey, A. (2008) A discussion paper on Robson Rotation in Tasmania, Tasmanian
Electoral Commission
http://www.electoral.tas.gov.au/pages/Media/PDF/Robson_Rotation_Paper.pdf
(accessed 10 June 2015).
26
Hermann, M. & Shikano, S. (2014) Attractiveness and Facial Competence Bias Face-
Based Inferences of Candidate Ideology. Political Psychology. (published online at
http://onlinelibrary.wiley.com/doi/10.1111/pops.12256/abstract)
Ho, D. E. & Imai, K. (2008) Estimating Causal Effects of Ballot Order from a
Randomized Natural Experiment, Public Opinion Quarterly, 72 (2): pp. 216-240.
Johns, R. & Shephard, M. (2011), ‘Facing the voters: The potential impact of printing
candidate photographs on ballot papers’, Political Studies, 59(3), pp. 636-58.
Kelley, J., and McAllister, I., (1984) Ballot Paper Cues and the Vote in Australia and
Britain: Alphabetical Voting, Sex and Title, Public Opinion Quarterly, 48 (2), pp.
452-466.
Kimball, D.C. & Kropf, M. (2005) Ballot Design and Unrecorded Votes on Paper-
Based Ballots, Public Opinion Quarterly, 69(4), pp. 508-529.
Koppel, J. & Steen, J. (2004) The Effects of Ballot Position on Election Outcomes”,
The Journal of Politics, 66 (1), pp. 267 - 281.
Lansdowne Market Research (2005) Ireland’s Invisible MEPs (Dublin, Lansdowne
Market Research).
Lansdowne Market Research, (1999) European Elections Ballot Paper Development,
(Dublin: Lansdowne Market Research).
Lansdowne Market Research, (2000) European Elections Ballot Paper Development
(Dublin, Lansdowne Market Research).
Lutz, G. (2010) First Come, First Served, The effect of ballot position on electoral
success in open ballot PR elections, Representation, 46 (2), pp. 167-181.
Marcinkiewicz, K. & Stegmaier, M. (2015) Ballot Position Effects Under Compulsory
and Optional Preferential-List PR Electoral Systems, Political Behavior, 37 (2), pp.
465-486.
Marsh, M. (1987) Electoral evaluations of Irish Candidates in Irish General Elections,
Irish Political Studies, 2, pp. 65-76.
Marsh, M. (2007) Candidates or Parties? Objects of Electoral Choice in Ireland, Party
Politics, 13 (4): pp. 500-527.
McDermott, M. (1998) Race and Gender Cues in Low-Information Elections,
Political Research Quarterly, 51 (4), pp. 895-918.
McDermott, M. (2005) Candidate Occupations and Voter Information Shortcuts,
Journal of Politics, 67 (1), pp. 201-219.
Meredith, M. & Salant, Y. (2013) On the Causes and Consequences of Ballot Order
Effects, Political Behavior, 35 (1) pp. 175-197.
Miller, J.M. & Krosnick, J.A. (1998) The Impact of Candidate Name Order on
Election Outcomes, Public Opinion Quarterly, 62 (3), pp. 291-330.
Murray, R, (2005) The power of sex and incumbency: a longitudinal study of
electoral performance in France, paper presented at the Political Studies Association
(UK) Conference, Leeds.
Norris, P., Vallence, E. & Lovenduvski, J. (1992) Do candidates make a difference?
Gender, Race, Ideology and Incumbency, Parliamentary Affairs, 45(4), pp. 496-517.
Oliviola, C.Y., Sussman A.B., Tsetsos K., Kang, O.E., & Todorov, A. (2012)
Republicans prefer Republican-looking leaders: Political facial stereotypes predict
candidate electoral success among right-leaning voters, Social Psychology and
Personality Science, 3 (5), pp. 605613.
O’Shaughnessy, N.J., (2004) Politics and Propaganda: weapons of mass seduction
(Manchester: Manchester University Press).
Ortega Villodres, C. (2003) Intra-party competition under preferential list systems: the
case of Finland, Representation, 4(1), pp. 55 66.
27
Ortega Villodres, C. & Garcia de la Puerta, B.M. (2004) Position Effects under STV:
Ireland and Malta. Representation, 41 (1), pp. 3 -14.
Pereira Montabes, J, & Villodres Ortega C. (2002) El voto en las elecciones al Senado
español: estrtegias de nominación y rendimientos partidistas en las elecciones de
marzo de 2000, Revista Española de Ciencia Politica, 7.
Pilet, J.B., Freire, A. & Costa, O. (2012) Ballot Structure, District Magnitude and
Constituency Orientation of MPs in Proportional Representation and Majority
Electoral Systems, Representation, 48 (4), pp. 359-372.
Popkin, Samuel L; (1991) The Reasoning Voter: Communication and Persuasion in
Presidential Campaigns (Chicago: University of Chicago Press).
Reif, K., & Schmitt, H. (1980) Nine Second-order National Elections. A Conceptual
Framework for the Analysis of European Election Results, European Journal of
Political Research, 8, pp. 3-44.
Reynolds, A. & Steenbergen, M. (2006) How the World Votes, The Political
Consequences of Ballot Design Innovation and Manipulation, Electoral Studies, 25
(3), pp. 570 598.
Robson, C, & Walsh, B., (1974) The Importance of Positional Voting Bias in the Irish
General Election of 1973, Political Studies, 22 (2), pp. 191-203.
Rosenberg, S.W., Bohan, L., McCafferty, P., & Harris, K. (1986) The Image and the
Vote: The Effect of Candidate Presentation on Voter Preference, American Journal of
Political Science, 30 (1), pp. 108 - 127.
Shephard, M. & Johns, R. (2008), ‘Candidate image and electoral preference in
Britain’, British Politics, 3(3), 324-49.
Sinnott, R., Coakley, J., O’Dowd, J. and MacBride, J. (2008) Preliminary Study on
the Establishment of an Electoral Commission in Ireland. (Dublin: Department of the
Environment, Heritage and Local Government).
Taebel, D.A. (1975) The Effect of Ballot Position on Electoral Success, American
Journal of Political Science, 19 (3), pp. 519-526.
Todorov, A., Mandisodza, A.N., Goren, A. & Hall, C.C. (2005). Inferences of
Competence from Faces Predict Election Outcomes. Science, 308 (5728), pp. 1623
1626.
... Specifically, good ballot design can help reduce the proportion of unintentional vote spoiling. The problem of invalid votes is important because the large proportions of unintentionally spoilt votes may affect the outcome of an election (Bonneau and Loepp, 2014;Gendźwiłł, 2015;Herrnson et al., 2012;Kortum, 2016;Kortum and Byrne, 2016;Pachón et al., 2017;Power and Roberts, 1995;Reidy and Buckley, 2015;Rusk, 1970;Seib, 2016). As Kortum and Byrne (2016) notice, democratic elections "must reflect the will of the voters. ...
... Badly designed ballot paper formats may undermine the electoral process and even change the election results. Also, Reidy and Buckley (2015) argues that the design of ballot papers sometimes "contravenes the assumption of electoral neutrality". ...
Preprint
Large proportions of spoilt or invalid votes may lead to a distortion of election results. While invalid votes are sometimes meaningful political acts, reflecting political protest, spoilt ballot papers may also be a consequence of a voter error, which can often be attributed to a poor design of a ballot paper. In this study, we use a natural experiment - employing empirical data from the Polish 2014 local elections - to verify the cause-and-effect relationship between the ballot paper format and the proportion of invalid votes. A substantial increase in the fraction of spoilt votes in the 2014 Polish local elections is commonly associated with the implementation of the new ballot paper format. We argue this is the first methodologically rigorous study testing the hypothesis that the ballot paper format redesign elicited an increase in the proportion of invalid votes in the 2014 Polish local elections. We apply the generalized synthetic control comparative method proposed by Yiqing Xu (2017) to show that the implementation of the complex or ambiguous ballot design can result in a substantial increase in the rate of spoilt votes. The paper demonstrates that the ballot paper design matters. Specifically, the results of this study confirm a cause-and-effect relationship between the ballot paper format and the fraction of invalid votes in the 2014 municipal elections in Poland.
... However, there is a growing body of literature on the effects of ballot design on voting behavior (e.g. Bonneau and Loepp, 2014;Carman, Mitchell, and Johns, 2008;Gendźwiłł et al., 2016;Herrnson et al., 2012;Kortum, 2016;Kortum and Byrne, 2016;Kimball and Kropf, 2005;Pachón, Carroll and Barragán, 2017;Reidy and Buckley, 2015;Reynolds, and Steenbergen, 2006;Roth, 1994;Rusk, 1970;Seib, 2016;Taylor, 2012Taylor, , 2017. For example, Reynolds and Steenbergen (2006) demonstrate that the so-called "elaborate ballots", which include photographs, colors, and symbols, may substantially affect a voter choice; they link the ballot design effect on voter behavior to theories about heuristic information-processing (e.g. ...
Full-text available
Article
Large proportions of spoilt ballots may lead to a distortion of election results. While invalid votes are sometimes meaningful political acts, reflecting political protest, spoilt ballot papers may also be a consequence of a voter error, which can be attributed to the poor design of a ballot paper. In this study, we focus on the widely publicized case of Polish local elections of 2014. These elections saw unprecedentedly large proportions of spoilt votes and ballot paper format was often mentioned as a potential culprit. Yet, it remains largely unclear whether or not there indeed does exist a cause-and-effect relationship between ballot card format and the tendency of voters to cast invalid votes. We argue that what we deliver here is the first methodologically refined test of the aforementioned hypothesis. We apply the generalized synthetic control comparative method, proposed by Xu (2017), and show that the implementation of a complex or ambiguous ballot design can result in a substantial increase in the rate of spoilt votes. We thus demonstrate that ballot paper design matters. Specifically, the results of this study corroborate the existence of a cause-and-effect relationship between ballot paper format and the fraction of invalid votes in the 2014 municipal elections in Poland.
... Badly designed ballot paper formats may undermine the electoral process and even change the election results. Also, Reidy and Buckley (2015) argues that the design of ballot papers sometimes "contravenes the assumption of electoral neutrality". As a result, some political parties may be in a more favorable position than other parties contesting elections. ...
Conference Paper
Large proportions of spoilt or invalid votes may lead to a distortion of election results. Invalid votes are sometimes meaningful political acts reflecting political protests of voters. However, in other cases, spoilt ballot papers are a consequence of a voter error, which can often be attributed to the bad ballot paper design. We use a natural experiment-employing empirical data from the Polish 2014 local elections-to test the cause-and-effect relationship between the ballot paper format and the proportion of invalid votes. We apply the generalized synthetic control comparative method proposed by Yiqing Xu (2017) to show that the implementation of the complex or ambiguous ballot design can result in the substantial increase in the rate of spoilt votes. The paper demonstrates that the ballot paper design matters-the results of this study confirm a cause-and-effect relationship between the ballot paper format and the fraction of invalid votes.
Article
Young people are under-represented in formal politics. While this may be a mere projection of their lack among voters and party members, the article investigates whether being young is a disadvantage in election processes, and if age effects differ by gender. Bridging the literature on gender & politics and political behavior, the article draws on an innovative sequential mixed-method design. Studying the 2019 Irish local elections, it uses 33 interviews to build hypotheses, which are subsequently tested on an original candidate-level dataset ( n = 1884). The findings suggest that, when controlling for party affiliation and political status, being young can provide a net electoral advantage to male candidates. In contrast, young female candidates appear to be advantaged by their age but penalized by their gender. The article thus contributes to our understanding about the conditions right at the start of political careers and the emergence of intersectional representational inequalities.
Full-text available
Article
The simultaneous scheme in the 2019 elections in Indonesia caused voter confusion, especially in the legislative elections. Citizens who are confused and disappointed when voting candidates characterize the declining quality of representation in electoral democracy. This study aims to determine the factors of confusion among citizens when they are in the voting booth. The case study research was conducted in Surabaya by interviewing 54 residents after the general election using recalling questions and in-depth interviews. The results show that the voter confusion factor is the effect of the complex design of the 2019 legislative election ballot paper, the lack of socialization about election procedures, and the difficulty of respondents understanding the simultaneous election models. The competency category shows that voter confusion is the respondents' low interest in political discussions and inadequate political knowledge. These two competence issues affect the quality of voters' political participation. Voter confusion in Surabaya generally motivates the phenomena of incorrect and misleading voting.AbstrakSkema serentak dalam pemilu 2019 di Indonesia menimbulkan fenomena kebingungan pemilih, khususnya pada pemilihan legislatif. Warga yang bingung dan kecewa pada saat memilih kandidat mencirikan menurunnya kualitas representasi dalam demokrasi elektoral. Penelitian ini bertujuan untuk mendapatkan faktor-faktor kebingungan warga saat berada di bilik suara. Penelitian studi kasus dilakukan di Surabaya dengan mewawancarai 54 warga pasca pemilihan umum dengan menggunakan teknik recalling question dan deep interview. Hasil penelitian menunjukkan bahwa faktor kebingungan pemilih merupakan efek dari desain kertas suara pemilihan legislatif 2019 yang kompleks, minimnya sosialisasi mengenai tata cara pemilu, dan sulitnya responden memahami pemilihan model serentak. Kategori kompetensi menunujukkan bahwa kebingungan terjadi karena rendahnya ketertarikan responden dalam diskusi politik dan rendahnya pengetahuan politik. Dua masalah komptensi ini berefek pada kualitas partisipasi politik pemilih. Kebingungan pemilih di Surabaya secara umum memotivasi fenomena incorrect voting dan misleading voting.
Full-text available
Article
Zusammenfassung Die vorliegende Arbeit nimmt die große empirische Vielfalt in der Gestaltung von Stimmzetteln zum Anlass, um sich mit Frage zu beschäftigen, welche Stimmzettelmerkmale von Wählerinnen und Wählern gewünscht werden. Dazu greift die Studie auf ein Discrete-Choice-Experiment im Rahmen eines Online-Surveys mit rund 1200 deutschen Teilnehmenden zurück. Im Experiment entscheiden sich die Teilnehmenden zwischen fiktiven Stimmzetteln, die sich im Hinblick auf die Informationen zum Wahlsystem, die Informationen zu den Kandidierenden und das Design unterschieden. Im Ergebnis zeigt sich, dass die Teilnehmenden einer zu ausführlichen, aber auch einer zu knappen Erläuterung des Wahlsystems kritisch gegenüberstehen. Weiterhin schätzen die Teilnehmenden die Angabe von Beruf und Alter der Kandidierenden, während die Angabe von Kandidatenadressen skeptischer bewertet wird. Zudem kann festgestellt werden, dass die Verständlichkeit das wichtigste Kriterium für die Bewertung von Stimmzetteln ist. Auf Grundlage der Befunde werden Merkmale geeigneter Stimmzettel herausgearbeitet und ein gelungener Stimmzettel vorgestellt.
Article
In this article, we provide an overview of the contributions to this special issue of Irish Political Studies on electoral management practices in Ireland. We document the rationale behind the issue and explore its contribution to our understanding of how elections are operated. We would like to thank the authors in this issue for participating in the Election Institutions and Electoral Integrity workshop at the National University of Ireland in October 2014. We acknowledge their valuable and generous comments on all of the articles in this special issue. The workshop was funded by a grant from the ‘New Foundations’ scheme of the Irish Research Council.
Full-text available
Article
There are two essential elements to this paper. In the first instance, we explore the specific details of revenue and expenditure trends for local authorities over the last decade. The analysis is framed against a longer term political context of forty which focuses especially on the weakness of local government in Ireland. Despite an official narrative of financial over dependence on central government, the comparative examination of budgetary records of local authorities reveals considerable diversity in both the revenue and expenditure patterns of authorities across the state. While some authorities are heavily reliant on central government funding, others have a much stronger base of local funding and indeed, the financial crisis since 2008 may have increased these differences. The second dimension to the research is an exploration of the impact of the great recession from 2008 on local government finance in Ireland. Using a framework of new institutionalism, we identify the crisis as another critical moment for local government. We consider the political, economic and administrative variables which have brought local government to a financial crossroads and we explore the potential for long lasting financial change in local government as well as speculating on what the nature and outcome of that change might be. 3
Full-text available
Article
It is generally assumed that a political candidate's appearance and style have an impact on voters and the choices they make on election day. Little research, however, has been done to investigate this claim. Here, the authors examine the role that these nonverbal aspects of candidate presentation play in the process of political communication. In the course of two related studies, the impact of candidate photographs appearing on campaign flyers is assessed. The results suggest that these photographs influence voters' perceptions of a candidate and this significantly affects their vote.
Full-text available
Article
How do voters in preferential-list PR (PLPR) systems make their candidate choices after selecting their party? We study this question in Poland and the Czech Republic, which use PLPR with different rules to elect the lower house of their national parliaments. In Poland, the system obligates voters to select one candidate from their preferred party, while in the Czech Republic voters have the option of casting up to 4 preference votes for candidates on the party list. Drawing on the ballot cues literature, we test for ballot position and rank effects in candidate preference vote shares. Further, we exploit the difference in voting rules to test whether ballot position effects are stronger in Poland where voters must cast a preference vote compared to the Czech Republic where preference voting is optional. We find that while being listed first or last has a disproportionately positive effect on preference vote shares compared to what would be expected based on ballot ranking in both countries, the ballot position and rank effects are much more powerful in Poland than the Czech Republic.
Full-text available
Article
Many studies have been devoted to the way electoral institutions shape the representative process in general, and of the relation between legislators and their local constituency. Yet, the majority of these have been hardly comparative. And when they compared countries, these countries rarely differed in many aspects of the electoral system. This article takes a different perspective by looking at the effect of the electoral formula, district magnitude and ballot structure on how important the constituency is for MPs in three countries that differ on these three components of the electoral system: Belgium, France and Portugal. And the results, though they should be confirmed by analyses on more countries, provide interesting insights. In particular they show that the difference in nature between single and multi-member districts is more important for constituency-orientation than differences between multi-member districts. Second, the article shows that the effect of electoral institutions is not straightforward: its impact is stronger on attitudes towards the constituency than on actaul behaviours of legislators regarding their constituency.
Article
On Thursday 30 May 2013 the Minister for the Environment, Community and Local Government, Phil Hogan, issued the Local Electoral Area Boundary Report which redrew the constituency boundaries for the Irish local government elections scheduled to take place in May 2014. The report was drawn up by an independent commission, of which the author was a member, over a period of six months from December 2012 to May 2013 and presented to the minister who promptly published the committee's findings. The report entailed the largest redrawing of local electoral area boundaries since the foundation of the state and involved significant reductions in both the number of councils and councillors. This article presents an insider analytical account of the boundary committee's work and ultimate report. It discusses the place of local government in Irish politics, the political context in which the committee was established, the reform of local government structures by the Fine Gael-Labour coalition government, and the ultimate redistricting of the boundaries. It also assesses the implications of the committee's report, and reactions to it, for both electoral integrity and electoral management in the Irish state.
Article
The process of constituency boundary revision in Ireland, designed to satisfy what is perceived as a rigid requirement that a uniform deputy-population ratio be maintained across constituencies, has traditionally consumed a great deal of the time of politicians and officials. For almost two decades after a High Court ruling in 1961, the process was a political one, was highly contentious, and was marked by serious allegations of ministerial gerrymandering. The introduction in 1979 of constituency commissions made up of officials neutralised, for the most part, charges that the system had become too politicised, but it continued the process of micro-management of constituency boundaries. This article suggests that the continuing problems caused by this system – notably, the permanently changing nature of constituency boundaries and resulting difficulties of geographical identification – could be resolved by reversion to the procedure that is normal in proportional representation systems: periodic post-census allocation of seats to constituencies whose boundaries are based on those of recognised local government units and which are stable over time. This reform, replacing the principle of redistricting by the principle of reapportionment, would result in more recognisable constituencies, more predictable boundary trajectories over time, and a more efficient, fairer, and speedier process of revision.
Article
Although research demonstrates that favorable ballot position can deliver candidates a small windfall of votes in local, nonpartisan, and primary elections, it is not clear whether ballot order laws have had any impact on the composition of U.S. legislatures. In this article, I estimate the substantive significance of ballot order rules by comparing the legislators of states that alphabetically order ballots to those elected by states that randomize or rotate ballot order. I also compare legislators elected by states that started or stopped alphabetically ordering ballots in recent decades. I find that states that alphabetically order ballots disproportionately elect candidates with early alphabet surnames. My research challenges the prevailing belief that ballot order affects only minor elections and suggests that seemingly innocuous rules have altered our political landscape. I conclude that arbitrary ballot ordering rules should be reformed to remedy their substantial impact on political representation.