Alphabetised Ballots; Bad for Democracy?

A debate has broken out on social media about candidates and where they come on the ballot paper. Kate Nicholl writes for us explaining the problem and why it can matter on polling day

I always said if I ran for election again I’d change my surname to something beginning with an A or a B. Nicholl is such an unfortunate political surname, especially in Northern Ireland where it gets lost after all the M’s. Because it is a truth generally acknowledged that candidates who are first on the ballot tend to receive a vote advantage compared to those whose names begin with letters later in the alphabet.

Emma Little Pengelly (also known as Emma Pengelly), was in the Newsletter today. There has been speculation (mainly on twitter) that Emma Pengelly is prefixing her surname with “Little” to be above Ruth Patterson on the Ballot, something the DUP MLA says is untrue.

But even if she had simply revived her previous surname just to give her an alphabetical advantage in May’s election, I’m not sure I’d blame her because I think alphabetical ballots are seriously unfair.

During the 2014 European election some Chinese constituents who couldn’t read English wanted to know how to vote for Anna Lo – a volunteer thought the ballot would be alphabetised according to Party and told them Anna (being the Alliance Party candidate) would be the name at the very top. I’m going to go out on a limb here and guess that was the first time Jim Allister got so much support from the Chinese community…

But there’s a serious question: how much do alphabetical ballots skew democracy? And what do we do about it?

A 2012 Parliamentary Affairs article looked at local council elections in Britain and the electoral advantage gained from a candidate’s surname being higher up the ballot paper. For ‘low-information’ elections (where voters don’t know or care enough to consider all their options) especially candidates located at or near the top of the ballot paper are advantaged relative to their competitors.

Research which analysed the names of members of Congress from 1949 to 2012 and state legislators from 1967 to 2010, found that alphabetical ballot lists account for the results in 10 Congressional seats and more than 70 seats in state legislatures. The author argues that alphabetically ordered ballots undermine the principle of fair elections, and that reform of ballot order rules is needed.

Measures could be taken to remove the effects of alphabetic bias, like randomising name order or candidates could be grouped by party on the ballot paper. Another option could be rotating the order of candidates’ names on ballot papers, i.e. ballot papers are printed and collated so that consecutive ballot papers do not show candidate names in the same order. In Tasmania they use a system called ‘Robson Rotation’, where the number of versions of the ballot paper is equal to the number of candidates and each candidate’s name therefore appears an equal number of times at the top and bottom of the ballot paper.

Is it time we addressed the issue of alphabetised ballots? And if so, what method would be best to adopt? And finally, would it make much of a difference given our propensity for tribal voting?

In the meantime, if anyone has any ideas for catchy surnames (preferably beginning with “A”), I’m open to suggestions.

, , ,

  • Belfast Barman(ager)

    Interesting idea – I take your point that research shows there is a true influence on election prospects based on alphabetising… it does make me question the very existence of democracy though. If the electorate vote based upon who is at the top of the ballot paper… perhaps as a society, we don’t deserve good representation.

    It certainly fits with the football chant, “We all dream of an assembly of Agnews”.

  • Kate Nicholl

    Chicken/egg argument regarding political engagement I guess, but important to remember alphabet can influence outcome, but certainly isn’t the sole factor. And we are *always* deserving of good representation 🙂

  • Jag

    Returning to reality, what is the evidence to support the statement above ” it is a truth generally acknowledged that candidates who are first on the ballot tend to receive a vote advantage” Are voters that thick that they can’t look down a list of 5-10 candidates?

  • Greenflag 2

    Ahern once very popular , Adams has been known to win a few , Aylward has made a comeback in Kilkenny.

    Heres a link which might give you some ideas .

    They are all here from Adams to believe it or not Varadkar and Zappone with a Neville and a Noonan in there somewhere .

    As to your point . The Tasmanian example you mention sounds fairer . And in close counts it probably could be proved that being at the top end alphabetically does give an advantage certainly for the lesser known candidates .

  • Greenflag 2

    In multi seat constituencies with two or three candidates from the same party running an Adams is going to do better than a Munster unless voters are ‘ managed and an Allen better than a Weir .

    In FPTP system as at Westminster it’s probably much less critical and the majorities are much larger numerically between winning candidates and second placers (losers ) . You can come second or third or even fourth or fifth in terms of first preferences under PR and get elected to a 3 seater if your name begins with a B rather than an M or a P most likely . Anyway there is enough of a bias under PR to shuffle the name lists .

  • Cavehill

    The major impact is when voters are giving their preferences. You go into the booth determined to vote Williams, but after that, what do they do? Well you go back to the top and look down to see who you give your second/third/fourth preference to – you’re more likely to plump for Adamson than Young.

    Names at the top of the ballot are more transfer friendly, by random luck of a aurname.

  • mjh

    In 2014 NI21 was a new party fighting its first election. None of its candidates had previous political records. In 8 District Electoral Areas they stood two candidates.

    And in every single case the candidate who came first on the ballot paper received more votes.

  • Greenflag 2

    Actually more likely to give it to either Adamson or Young instead of Mallon .Either end of the Ballot sheet is preferable to being in the middle ,

  • Msiegnaro

    That’s the stupidity of the electorate if candidates are being voted for based on alphabetical positioning.

  • mjh

    Not stupidity. Those people chose to vote for a particular party. Based on the information at their disposal they appear to have seen no significant difference between the candidates and so voted “1” for the first of the party’s candidates on the paper and “2” for the second. Seems like a perfectly natural thing to do.

    Or do you mean that the clever ones started at the bottom?

  • Msiegnaro

    If one is going to vote it would seem logical to do one’s homework on the candidates in advance.

  • SeaanUiNeill

    Not thick Jag, just without any real interest in the outcome…..

  • Greenflag 2

    Some voters just vote the party and don’t care whether the candidates name is Adams , Moore or Zablonski . As we are accustomed to reading lists of names top to bottom in normal daily life or lists generally most people start at the top and work their way to the bottom . They may miss Moore on the way down and instead tick Monroe or Molyneaux and not notice it . They’ll then give preferences to other parties or personalities that they favour and sttrt going up or down the list again . Its the way the brain works for most of us . When there are 10 or more candidates this up and down tends to favour thse at the top or bottom not in all cases but probably in enough to be significant to the result when the seat is decided by a couple of hundred votes or so .

    The highly motivated voter who knows how to maximise his or her vote will get it right . Many voters will get the first 5 or so preferences accurate and then start to avoid the names or parties they don’t want to see elected and then they’ll tick off the remaining names from top down or bottom up mostly .

    Somehow it works to ensure that the candidates with the most first preference votes got elected to all but 12 of the 158 Dail seats in GE # 16 and I think the same will apply in the Northern Ireland Assembly election in May . It’s probably a good research project for somebody studying comparative electoral systems in western democracies or some such .

  • Greenflag 2

    Yes-people should be sent a sample copy of the actual ballot a couple of weeks before election day or have it e-mailed to them so they can fill it in without pressure and then just copy it to the real ballot in the polling booth .Could speed up voting process and reduce lines not that they are a big problem given almost half of the electorate won’t even get to the polls:(..

  • Reader

    Probably more of an issue within parties than between parties. For instance, a DUP voter might work down the DUP candidates in order, then think about their next party to hunt for in the list. Surely almost no-one will walk into a polling station without first having picked a party for their top preferences?

  • Msiegnaro

    That already happens.

  • Greenflag 2

    Well then send another just in case the first is mislaid or e-mail . You can lead a horse etc but it is up to people to whether they drink or vote or not .

  • Msiegnaro

    For goodness sake, you’re asserting that the electorate are incapable of making informed decisions by themselves and instead seek to spoon feed them on everything.
    I despair of the people whom you are referring to having a vote.

  • Greenflag 2

    And I sometimes despair of some of the people they vote for . Churchill remarked that you only had to talk to the first ten people you meet in the street on any political and you would know that democracy could’nt work .
    He also added much later that even so it was an improvement on all the alternatives tried in history . And he’s still right on that score .

    While some of the electorate may be incapable of making informed decisions this is not the case for all or even most of the electorate . Enough people do vote and know what they’re voting for or against to keep the show on road . And thats what matters for if the show goes off the road it can lead to the Gulag or a Dachau or the killing fields where votes don’t exist and insanity reigns over human destruction .

  • Msiegnaro

    I agree with what you’re saying with the last point, however before that you claimed that unless the electorate knew the alphabetical order of their preferred candidate they would be disenfranchised.

  • Jag

    As far as I can see Greegflag2, there were just 168 out of 550 candidates whose surnames were N-Z.

    So, 31% of candidates were N-Z and 8% of elected TDs were N-Z.

    That seems to assist the claim that names in the A-M range are more successful. Would be interesting to do analysis of the original candidates for selection to see if the effect is further magnified.

    Still can’t believe Emma would be that petty though, and her Twitter has always been Little_Pengelly.

  • Jag

    Might be right Seaan, but if you’ve bothered to register to vote, and take the time to arrange to visit the polling booth to cast your vote, you’d think that you’d pay more attention to your vote.

  • Greenflag 2

    ‘to see if the effect is further magnified.’ It probably is . I guess there’s also the total of number of surnames in a population and what percentage of the total are in the A to M range and the N to Z range . Perhaps 80% of names are in the A to M range ? and only 20% in the N to Z range which using your data above would mean the N to Z group are overepresented as candidates but underrepresented as elected TD’s . As for Emma being petty ? I can’t comment but sometimes you’ll find double barreled names used by candidates where the first letter of the first part of the name is a lot closer to the beginning of the alphabet than the end .

    Of course already elected TDs or MLA’s won’t want any change . Are British Westminster election lists also alphabetically based do you know ?

  • Greenflag 2

    Disenfranchised ? Can’t recall using the word not ever . They may vote for their party candidates in alphabetical order which may not be the best order to win say 2 out of 3 seats in a 3 seat constituency or three in a 5 seater . With FPTP theres no thought needed just the 1 vote one time and if the piggies name is Porky as long as she’s wearing a blue ribbon she gets your vote or alternatively as long as the other piggies name is Boris and he’s wearing a red ribbon he gets your vote . Its as they say a no brainer . PR requires thought and can be challenging for some .

  • mjh

    Yes they are.

  • mjh

    The alphabet effect may only apply within parties and in the least important elections where candidates may not be so well known.

    In the 2011 Assembly elections 50.5% of the successful candidates came from the top half of the ballot paper and 49.5% from the bottom half.

  • Gavin Neville Charles Morrison

    If anyone wants to have a look at an academic treatment of the use of randomisation in democratic decision making procedures then the book ‘The Luck of the Draw’ by Dr. Peter Stone is reasonably exhaustive (and a surprisingly good read). If you don’t want to read the book his argument, in short, is that using randomisation can have a sanitizing effect on making decisions- in a scenario in which there is no good criterion upon which to base a decision (i.e. only bad reasons to prefer option a to option b) then randomisation can sanitize the decision making process- by taking all reasons out of the decision. As for the collective irrationality of most electorates…

  • SeaanUiNeill

    You’d think so, but all too many of our fellow citizens seem to be driven by one issue voting such as Arlene has recently thought to stimulate. Hardly “informed choice”…

  • chrisjones2

    You assume that all voters can even read and write