Over 40,000 e-petitions have been submitted to parliament since the current system was introduced in 2015. Cristina Leston Bandeira and Viktoria Spaiser have conducted research into how the public views the consequent parliamentary discussion of issues raised in these petitions by analysing comments made by those watching the live parliamentary coverage. Their findings lead them to conclude that parliamentary debates should be adapted to be more inclusive of the original petitions’ aims.

Parliament introduced an e-petitions system in 2015 with the aim of enhancing its relationship with the public. The system has seen extraordinary levels of usage, with over 40,000 e-petitions submitted and plenty of other evidence of very considerable engagement from the public, such as petitions debates regularly being the most read debates on Hansard. The extraordinary usage is only one element of this new system, however. At the Centre for Democratic Engagement, we have been investigating it, focusing in particular on the more subtle expressions of engagement, beyond usage numbers. We have interviewed petitioners, developed participant observation, and analysed petitions data, parliamentary documentation and social media activity associated with e-petitions.

Some of this research has now started to come out, namely our latest article in Policy & Internet, where we use natural language processing, machine learning and social network analysis of Twitter data to explore what it shows about the extent of people’s engagement, the contents of Twitter e-petition conversations, who is taking part and how they interact. In this blog post we focus on how the public react to the format of the e-petitions parliamentary debates, through their comments on Twitter whilst they watch these debates. Our findings provide interesting insights into how people perceive the e-petition procedures in terms of fairness and responsiveness, suggesting that petition parliamentary debates could be more inclusive of the original petitions’ aims.

The new parliament and government collaborative e-petitions system is a markedly new system in relation to the preceding widely criticised government-led e-petitions sites, crucially because it formally integrates parliament into the process. Amongst its key changes is a clearer channel for responses to e-petitions, with those reaching 100,000 signatures being considered by the Petitions Committee for a debate. This was in response to assertions that the previous, government-led system was failing to deliver on its promises, and was risking ‘an exacerbation of public disillusionment with the political system in the long-term’. Since then, the e-petitions system has mediated very significant activity and the Petitions Committee has developed a plethora of public engagement activities utilising diverse tools such as Twitter. However, we still know little about the extent to which the new system has enhanced the relationship between public and representatives, namely whether the new system promotes a better parliamentary process to consider the issues raised by petitions.

Using Twitter data to understand how the public react to parliamentary process

We utilise Twitter data to assess this issue. Seeing how important the parliamentary debate is as a key step in the petitions’ consideration process, we explore what Twitter conversations tell us about what people think of the debate. We don’t take Twitter data as representative of the view of the general public and we acknowledge that individuals engaged in Twitter discussions about a particular e-petition may not necessarily have signed it. Nevertheless, the data provides real-time reactions to a parliamentary event, giving a useful thermometer of how people interested in the issues of a specific petition react to a key moment in parliament’s consideration process.

This matters for a number of reasons, namely because the new e-petitions system was introduced with the intention of fostering a significant enhancement of the relationship between public and parliament. This implies the development of a perception that the institution listens to the public’s concerns expressed in an e-petition. The parliamentary debates of e-petitions with over 100,000 signatures are seen as the pinnacle of the system, but there has been little consideration as to what actually happens within the debates. Twitter discussions enable us to tap into real-time public reactions to how their petitions are being considered by parliament.

Our Twitter data relate to conversations on 33 parliamentary occasions associated with the 28 e-petitions that were granted a parliamentary debate between March 2016 and May 2017, including: 28 debates, four oral evidence hearings and the launch of one inquiry report. Although in the article we make an overall analysis of the data for these 33 cases, here we discuss only our in-depth analysis of a single e-petition calling for a ban on driven grouse shooting.

This makes an interesting case study for two reasons. Firstly, as the practice is both fiercely protected and strongly opposed, it refers to something specific sectors of the public feel passionately about, being more likely therefore to spark discussions. This e-petition generated very active discussions on Twitter, providing for rich data to enable deeper insights into how people react to a parliamentary session. Secondly, besides a debate, this e-petition also led to an oral evidence session. This enables us to compare public reactions to how e-petitions are addressed between two different types of parliamentary sessions. This matters, because perceptions of petitioning vary according to how its processes are judged by petitioners. The comparison of the public’s reaction to two different parliamentary processes helps us to better understand the extent to which the new system has enhanced the consideration of petitions. Although here we focus on the driven grouse shooting petition, our analysis of a very different type of e-petition (on giving meningitis B vaccine to all children) backs up the conclusions of our case study on grouse shooting.

When a petition parliamentary debate may not necessarily enhance engagement

Figure 1 (below) shows the topics covered in the Twitter discussions about our case study e-petition, by depicting its respective semantic network: this consists of words (nodes) and edges that connect words that appear close in a text, allowing us to visually represent meaning and themes (clusters within the network) in a text.  The network is split into topic clusters identifiable by their different colours. Among these clusters, several reflect clear criticism of the petition debate procedure. The yellow cluster references specifically the MP who led the debate and describes it as polarised, frustrating and biased. This reflects the criticism that far more time was given to the counter arguments opposing the ban (and therefore opposing the e-petition), than those in support of it. The focus of this Twitter conversation shifts from a discussion of the facts (the purple cluster constitutes the only discernible factual topic covering aspects of environmental impact) to a critique of the fairness of the debate procedure.

Figure 1. Semantic network based on tweets on the ‘Banning Grouse Shooting’ e-petition parliamentary debate (31 October, 2016).

In order to better assess these findings, we also constructed a semantic network for this petition’s oral evidence session, which preceded the parliamentary debate. Figure 2 (below) shows that this session’s Twitter conversation remained more factual, less polarised and with essentially no critique of the parliamentary process. Twitter was used more as a tool to relay information and facts, which were presented in parliament to a wider audience during the oral evidence session. It suggested, for instance, that more evidence is needed to fully assess the impact of grouse shooting. The orange cluster in Figure 2 illustrates this, with calls for more transparency about who owns grouse moors, and in the light blue cluster, which suggests more information is needed about the impact on the environment of burning heather moorland.

Figure 2. Semantic network based on tweets about the oral evidence session on the ‘Banning Grouse Shooting’ e-petition (18 October 2016).

Comparison of the Twitter conversations in Figures 1 and 2 suggests that, at least in this instance, the parliamentary debate may not be conducive to fostering trust in the process. There are clear differences in procedure between a parliamentary debate and an oral evidence session. The latter focuses on the interrogation by MPs of evidence and facts presented by witnesses. A parliamentary debate, however, is a discussion amongst MPs alone. Given that a large number of Conservative MPs (i.e. opponents of the grouse shooting ban), attended this specific debate, it quickly turned into a party political event. Furthermore, seeing the low numbers of opposition MPs attending the debate, it soon became mainly a critique of the main e-petition being discussed. The semantic network analysis in Figure 1 suggests that members of the public felt that this process took insufficient account of their views  and those of the main petitioner, and that it granted unfair weight and bias towards specific groups perceived to hold political favour. These perceptions are not limited to this e-petition, as we observed similar differences between the debate and the oral evidence hearing for the e-petition on the Meningitis B vaccine for instance.

What is more, our interviews with petitioners confirmed that they were generally very positive about the new petitioning system: the only element that sometimes disappointed was the actual debate. This was expressed by those who felt the debate did not reflect the intent of the originating petition, regardless of whether their petition had achieved anything in addition to securing the debate.

What does this tell us?

The debates analysed in our article are now over two years old and some evidence suggests that e-petitions debates have become more focused on their originating e-petitions since then. But our analysis suggests two important findings to better understand public reaction to how parliament considers e-petitions. Firstly, a well-attended parliamentary debate with considerable MP participation does not necessarily equal an outcome that it will be well received by the public. The voices expressed in the debate and the extent to which these reflect the intentions of the e-petition in question are far more important. For example, despite very low MP attendance, the debate on the April’s Law e-petition was far more inclusive of its originating e-petition. Petition debates tend to be conducted as any other parliamentary debate. Our analysis suggests that they should reflect more their distinct nature, which derives from the originating e-petitions, by for instance acting more as an advocate for the petition and focusing more explicitly on petitioners’ aims. This relates to a wider issue, which is how political institutions integrate the voice of the public into their processes. Bolting the public’s voice onto established practices seldom leads to a genuine representation of the public’s view.

Secondly, the public tends to react in a more polarised way to parliamentary debates than to oral evidence sessions. This will be of no surprise to those familiar with Westminster. Debates are conducted under long embedded traditions of adversarial politics. In contrast, oral evidence sessions, which take place in committee, tend to be conducted along consensual lines. The focus is on the witnesses and on their interrogation to establish facts and evidence. Our analysis of Twitter conversations confirms that the public reacts more negatively to parliamentary debates than to oral evidence sessions. One could argue that this is only natural and merely reflects the nature of these parliamentary procedures. However, the reaction on Twitter shows the extent of the frustration amongst the public caused by a debate that was overwhelmingly in opposition to its originating petition. These findings suggest that more care should be taken over the extent to which these debates reflect the original purpose of their petitions. It is worth noting that other parliamentary petition systems (such as in the European, German and Scottish parliaments) do not put as much focus on debates, prioritising instead evidence sessions with petitioners.

Our article also demonstrates the value of big data analysis to understand even the most traditional of processes, such as a parliamentary session, as they enable us to observe public reactions in real-time.

The full article, ‘Do Parliamentary Debates of e-Petitions Enhance Public Engagement with Parliament? An Analysis of Twitter Conversations’, is available here. Copies can also be obtained from the authors.

About the authors

Cristina Leston-Bandeira is Professor of Politics at the University of Leeds and co-director of the Centre for Democratic Engagement. She tweets as @estrangeirada.

Viktoria Spaiser is a University Academic Fellow in Political Science Informatics at the University of Leeds. She tweets as @viktoria_sp.