NI PoliGraph is a new website that provides ongoing data-driven analysis of Northern Ireland politics. The website tracks data from a variety of sources – Twitter, news outlets, the congregation, and polls – for NI political parties, as well as our MLAs and MPs.
What the page contains
Twitter is used to some extent by the majority of MLAs and MPs, and while there is a significant amount of “noise” in the content of members’ posts or retweets, we can still learn something by following this activity: Who is Using Twitter ? Most who have the greatest influence on how party members interact with each other and (to some extent) what is tweeted. Below you can see the most effective members on Twitter, measured by the average number of retweets per tweet in the last month (as of late November 2020).
The largest dataset on the website relates to NI Assembly activities: ministerial questions and answers, committee activities, transcripts of debates and votes, accessed through the AIMS portal. Natural language processing is used to keep track of the issues discussed in plenary sessions and statistics are calculated on how members and parties voted. This section includes the option to view previous congregation sessions up until 2007 to relive some forgotten names from the recent past.
Also included are mentions of the politicians in a number of online news sources and an opinion poll tracker that runs from 2015 to date.
Finally, this data is summarized in a “testimony” for each politician to put the member’s overall activity in relation to his colleagues in context: how effectively they use Twitter, how often they are mentioned in the media and how much and type of them they are Contributions to the assembly, including their voting results in the current session (for MLAs). This is partly due to similar functionality provided for Westminster by They Work For You and for the US Congress by GovTrack. The activity part of the assembly of the report can highlight a member’s preferred discussion topics and, to some extent, the tone of their contributions, as shown in this example (for Paul Frew MLA, late November 2020).
Voting results of the meeting
Analyzes are also occasionally carried out on the website using the various data collected.
In the first instance, the voting dates of the meeting from 2007 to today are used. One trend that we are seeing here is that the rate of “yes” votes has risen over this period, which means that motions are more likely to receive majority support. However, the need for cross-community support has also increased, resulting in only a small increase in the frequency of requests.
We can also examine how each individual MLA tended to vote based on whether the motion was proposed by unionist or nationalist members (excluding motions from members labeled “Other” or a mix of labels). In the illustration below, items in the lower right hand corner represent members who normally voted for a unionist motion and voted against nationalist motions. Points in the top left represent the opposite scenario. Some interesting outlier MLAs are highlighted.
The nomination of the applicant is clearly a strong indicator of how a member will vote on the application, but there is a spectrum of strict adherence to this tribal practice. Alliance members sit at the center of the conspiracy for the most part, which means they have supported both unionist and nationalist proposals roughly half the time. There are some members and parties – those in the upper right corner of the property – who show a greater tendency than others to vote for a motion in general, regardless of whether it is proposed by members of the same or a different designation has been .
The full article is available from NI PoliGraph.
What NI PoliGraph is for
Specifically through the individual membership reports, I hope this website can improve the scrutiny of our elected representatives by allowing readers to focus on the actual performance of their MLA rather than just looking for party labels. It could be useful reference material for “traditional” political journalism as featured in Slugger O’Toole. At the party level, the voting papers and social media data can clarify what each party “means” and how they relate to each other. This is particularly useful information in the context of the ranking voting system we use in NI for general elections. After all, it’s for people like me who enjoy looking at graphs and are curious about long and short-term trends in politics that data might reveal.