One of the areas that my team at the Electoral Commission is responsible for is leading our work registering political parties in Great Britain and Northern Ireland. Registering as a political party allows that party to use particular ‘identity marks’ on the ballot paper. These identity marks include the party name, up to 12 descriptions (which can appear instead of or alongside the party name depending on the election) and up to 3 emblems.
This is an important area of work for the Commission. Shortly before we were established, a case involving the formation of a party called ‘The Literal Democrats’ led to the Registration of Political Parties Act 1998, which outlawed the use of names in elections designed to cause confusion with more established parties. The responsibility for overseeing this area passed to the Commission shortly after it was established by the Political Parties, Elections and Referendums Act (PPERA) in 2000.
As Head of Research at the Commission, another area that my team leads on is making sure that the work we do, and the decisions and recommendations for change we make, are as informed by voters and their views as much as possible. Today (6 August), we’ve published research on how voters find the political party or candidate they wish to vote for on their ballot paper.
When we receive an application, either from a new party or from an existing registered party wishing to change its details, we need to consider the particular identifiers applied for against some legal tests set by PPERA.
Some of these tests are straightforward, for example an identifier cannot be registered if it is the same as one for another registered party or if it contains more than six words. However, others are less specifically defined within the legislation, which gives the Commission a wider discretion to form an opinion on whether each identifier meets each test or not. For example, an identifier cannot be registered if it is likely to result in voters confusing it with another identifier that is already registered. It can also be a challenging area of work with 496 parties registered across Great Britain and Northern Ireland. These parties have a total of nearly 1,500 descriptions and over 650 emblems registered with the Commission.
The research we’ve published today was designed to provide us with an up to date evidence base on how the similarity of party identifiers could affect a voter’s ability to vote according to their intention and, potentially, to highlight any changes we needed to incorporate into our approach to assessments. The research was carried out for the Commission by Nat Cen who used cognitive interviewing techniques to explore how people interact with the ballot paper. Each participant was given a mocked up ballot paper (containing a mix of real and fictional parties) and asked to cast a vote for a specified party. They were then asked about how they had gone about finding that party on the ballot paper.
Overall, the key findings from the research confirm our current approach to assessing applications. In particular we found that:
- Voters use mainly party name and emblems (and the candidate’s name) to identify their chosen party, with many looking at a second identifier to double-check.
- A single similar identifier can cause confusion and this should be taken into account in making assessments but…
- …there is an additional effect from having similar names and emblems which exacerbates confusion.
- A single geographical identifying word (e.g. ‘London Conservatives’ or ‘Cardiff Conservatives’) is not sufficient to distinguish between parties, and names/descriptions should have further distinctive elements to avoid confusion.
- Overlapping ‘ideological’ words can be problematic (more than other non-ideological words) – e.g. Liberal, Conservative or Socialist in a name needs to be clearly distinguished from other party names and descriptions using the same word.
- Our assessment of emblems needs to take account of the pictorial and textual elements – e.g. two emblems with the same text, written in a similar style, can be confusing even if the images are different.
- Our assessment of emblems needs to consider the extent to which the objects in an emblem are depicted in a visually similar way, rather than simply the fact that two or more emblems depict the same object – e.g., two trees may be depicted in very different ways.
While these findings already formed part of our approach to assessment there was value in confirming, with research evidence, that our current way of doing things is valid.
The Commission keeps its approach to assessing applications to register as a political party under regular review. We have also made a number of recommendations to government for changes to aspects of the legislative framework in our recent report on the May 2015 elections. In particular, we are concerned that the legal provisions for registration of party descriptions (which do not have to include the party’s name) present a risk of confusion for voters and restrict the participation of political parties. We continue to recommend that the UK Government should, therefore, reform or remove the provisions on party descriptions.
In the meantime, we will also consider whether there is any further research we can do to make sure our process for making decisions, and the decisions themselves, are as robust as possible. For instance, this particular research did not include any fieldwork in Northern Ireland. Although we have no reason to believe there are significant differences in people’s understanding of the ballot paper in Northern Ireland compared to Great Britain, this may be another area where future research would be useful to test our assumptions. We’ll use the blog to keep you up to date with any future research we publish and in the meantime do let us know if you have any comments.
Head of Research and Party Registration