Debating the TV Debates

By Darren Hughes

In an era of political disillusionment, finding ways to bring politicians closer to the public is not just desirable but essential for the health of our democracy.

The ERS supports TV debates as a way of getting election debates and key policy issues heard. And not just on a UK-wide level; we were instrumental in ensuring the 2016 Assembly election debates in Wales featured the six key parties there.

In May 2017, amid suggestion that neither Theresa May nor Jeremy Corbyn would be appearing in a live debate, ERS research found that 56 percent of people believed leaders’ debates were important in helping them make their decision. More importantly, we found that the vast majority of those with a view believe that “all major party leaders should commit to participating in televised General Election debates”. What this research suggested was that voters take the TV debates seriously – and they want party leaders to take them seriously too.

June 2017’s snap election didn’t see a full, head-to-head debate between the two main party leaders (although there was a live debate in which the Prime Minister did not take part). The BBC’s Question Time leaders’ special was the closest we got and it too proved popular. More than four million people tuned in to see the main party leaders pitch their case and be challenged by a live audience.

In the run up to the programme, ERS commissioned leading academics in the field of communications and media to look at the impact of the debates on viewers. This research revealed how well TV debates deliver on citizens’ expectations of political communication (‘entitlements’ in the study). These five expectations are that leaders ‘put their points clearly’, ‘provide factual evidence’ and ‘a clear choice’, ‘engage me in the debate’, and ‘understand people like me’. The report finds that viewers expected, and leaders’ delivered on, most of these expectations in the Question Time special.

Over a third of viewers said the Question Time election special influenced their vote. On a UK-wide level, that would amount to 1.4m voters. These figures matter when ERS research shows the Conservatives could have won an overall majority with just 533 extra votes in the nine most marginal constituencies, while a working majority could have been achieved on just 75 additional votes in the right places.

That means small factors can have a significant effect on elections – a reflection of a broken voting system which needs replacing. But televised election debates are good for our democracy, as this report shows.

Over 80 percent of viewers said they talked about the Question Time special with their friends and family, while 40 percent said the programme made them more interested in
the campaign.

It was the ‘youth surge’ this election that was arguably most significant. As the authors reveal, more young people watched the entirety of the programme than older viewers, with a much higher proportion of young viewers undecided on who to vote for before seeing the Question Time special.

TV debates have become incredibly important for General Elections in the UK. And the positive democratic legacy of the BBC’s leadership special means it’s time to make such debates a core and established part of 21st century campaigning in the UK – with party leaders expected to take part. As this report concludes, we need to ensure citizens are ‘addressed, informed, engaged, recognised and empowered’ in ways that enable them to fully engage as a democracy.

The ERS is calling for a framework to be put in place so that live debates are fixed as an integral part of election campaigns. And so that any such programmes should be real head-to-head debates, open to meaningful and live challenge from opponents.

This report sets out the major impact of June’s leadership special for the first time. Now it’s time for party leaders and broadcasters to learn from voters’ views – and ensure that the debates are even better next time.

Darren Hughes
Chief Executive

The Question Time Special

That voters deserve more and better from democracy has become something of a commonplace observation – both from professional pundits who despair at the evasions, simplifications and over-assertions of politicians, and from citizens who are increasingly fed up with the entire political class.

But while such criticism is sometimes reduced to a ‘they’re all as bad as each other’ dismissal of political representatives, it should be taken very seriously.

Firstly, representative democracy cannot possibly work without representatives who are regarded with a degree of trust and without a represented citizenry that feels capable of translating its will into policy through peaceful means.

Secondly, when people give up on politics, it doesn’t go away but is left to the machinations of ‘insiders’ who are driven by narrow experience and interests.

Thirdly, democracy is the most creative and civilised way to run a mature, inclusive society, and any alternative regime is likely to be much worse at reflecting what people really want.

At the core of contemporary tensions between elites and citizens lie not just perceived failures to deliver political goods, but in particular, failures of communication. There is no shortage of ways that citizens are addressed by politicians via the mass media (television, radio and the press) and social media (from blogs and YouTube to Facebook and Twitter), but many of the priorities, styles, techniques and values of such address are not working.

In this study, we attempt to understand more about how citizens would like to communicate with politicians and how they want to be addressed by them. The Electoral Reform Society’s review of the UK’s 2016 European Union Referendum campaign was entitled It’s Good to Talk. But in discourse relations between politicians and voters, what kind of talk is ‘good’ and what ‘bad’?

Spanning the election campaigns of 2015 and 2017, our research has zeroed in on televised election debates as a primary example of ‘politician speak’. We know from our previous studies (Coleman, 2011; Coleman et al 2015) that televised election debates reach parts of the electorate that no other aspect of campaigns can.

Several months before the start of the 2015 campaign, we organised a series of focus groups, in which we asked a varied range of voters and non-voters to reflect on their experience of watching or having heard about the debates that were held in 2010 and then tell us what they hoped to gain from future debates (Coleman and Moss, 2016).

On the basis of what they told us, we identified five ways they thought that debates could help them to perform their roles as democratic citizens.  Each of these ways related to communication. In the body of this report, we term
them ‘entitlements’¹.

• To play their part in democracy as capable citizens, people felt entitled to be treated by politicians as follows:
• They wanted to be addressed as if they were rational and independent decision-takers
• They wanted to be able to evaluate the claims made by debaters in order to make an informed voting decision
• They wanted to feel that they were in some way involved in the debate and spoken to by the debaters in that spirit
• They wanted to be recognised by the leaders who claimed to speak for (represent) them
• They wanted to be able to make a difference to what happens in the political world

In follow-up surveys, these desired outcomes were put to respondents, asking them beforehand how confident they were that the leaders taking part in debates would:

• Put their points across in a clear, understandable way
• Provide factual evidence to back up the points they make
• Engage me in the debate
• Prove that they understood people like me
• Provide them with clear choices to vote for

And those who had watched a debate were asked to what extent they felt that the leaders had spoken in these ways. Following our 2015 findings – based on the seven-leader debate organised by ITV – we decided to conduct a similar study in 2017.

This time it was based on viewers’ responses to the BBC’s  Question Time Special, broadcast on 1st June, in which Theresa May and Jeremy Corbyn separately answered questions from members of a studio audience.

We commissioned ComRes to recruit a panel of 2,500 individuals who were questioned before and after the programme. The panel was nationally representative for age, gender, social grade and region. Our objective was to understand this important political moment within the context of that sense of estrangement from the ‘political class’ that had hitherto seemed so pervasive.

The electorate and the Question Time Special

Interest in and engagement with the BBC’s Question Time Special among our panel members was on balance relatively high. For example, when asked before its transmission whether they thought they would watch it, about a quarter said they ‘definitely’ would do so, a third that they ‘probably’ would, a fifth ‘probably not’, and a sixth ‘definitely not’.

Levels of political trust had apparently shaped these intentions strongly. Having been asked, ‘To what extent, if at all, do you trust politicians to do what is right for the country’², the more trusting respondents were more likely to view the programme. Three quarters of the ‘definite’ would-be viewers had at least some trust in politicians compared with two fifths of those definitely not intending to see it.  Put the other way round, 54 percent of those who definitely would not watch it had no trust in politicians, compared with 39 percent of the ‘probably not’ panel members, 20 percent of the ‘probably would’ and 22 percent of the ‘definitely would’ viewers.

Even so, readiness to view the programme was not confined to political cognoscenti. Among voters who described themselves as only ‘fairly interested’ in politics generally, 62 percent expected to view it.

In the event, two fifths of the respondents did tune in to the Question Time Special on the night, a half of whom claimed to have seen the full 90 minutes of it. In fact, most of the viewers (84 %) found it ‘good to talk about’ the programme with someone during or after it. Viewers discussed the programme most often with a partner (42%), family member (36%) or with friends (22%), and the bulk of these engaged in face-to-face rather than online conversations (88%).

For many viewers, watching the programme was also something of a learning experience.  For example, four fifths of them declared that they had been ‘able to understand the claims and arguments put forward by both sides’ (rather than ‘unable’ to do so). Half felt that they now knew ‘more about some of the Conservative and Labour policies than I did before the programme’ (rather than ‘I did not learn anything new about Conservative and Labour policies’).  And a similar proportion (46%) felt that after watching the programme they knew ‘more about what Jeremy Corbyn and Theresa May are like as people’.

Exposure to the Question Time Special may also have had important pay-off consequences. When asked about their interest in the election campaign ‘as a result of watching the programme’, as many as 40 percent of  the respondents said they had become ‘more interested’ in it, compared with 59 percent neither more nor less interested and only six per cent less interested.   

Thirty per cent of respondents endorsed the statement that ‘The programme helped me to decide whether to vote on Polling Day’ and 34 percent considered that ‘The programme helped me to decide what party to vote for’. Moreover, people who were less certain about their voting intentions were particularly inclined to say that viewing the programme had helped them to decide whether to vote (38% of the ‘fairly certain’ vote intenders and 44 percent of the uncertain ones, as well as which party to vote for (37% and 44% respectively).

But what about those discourse entitlements with which our research was centrally concerned? Did people expect the party leaders to address them well as democratic citizens? And did they find that they had answered the questions put to them by the Question Time studio audience in ways that would help or hinder democratic reflection?

Five main points emerge from the empirical evidence about this.  First, both their prior expectations and their after-viewing evaluations were modestly positive. Secondly, there were only minor signs that watching the Question Time Special had changed viewers’ evaluations of the leaders’ modes of address. Nevertheless, thirdly, a degree of prior confidence in how the leaders would speak seems to have encouraged some of them to go on subsequently to watch the entirety of the programme.

Fourthly, in interesting ways, ratings of the leaders’ modes of address differed to some extent across the five entitlement measures. And lastly, Jeremy Corbyn was regarded as having done a better job of speaking to them as they would prefer than did Theresa May.  The bases of these generalisations can be seen in the following tables.

Table I

Table II

Table III

During the debate, which of the leaders, if any, was most effective in:

Thus according to Table I, for four of the five measures, the greatest number of respondents intending to watch the Question Time Special were ‘fairly confident’ that the leaders would deliver the ‘entitlement goods’ (very few said that they were ‘not at all confident’ about this.)

The stand-out exception to this pattern was the response to ‘understanding people like me’, a majority having said that they were ‘not very confident’ or ‘not at all confident’ about the leaders’ ability to prove themselves in that regard. This may reflect a core element in public disenchantment with the country’s political elite today.

A similar but not entirely identical pattern appears in the right-hand columns of the table, showing how the viewers of the Question Time Special assessed the leaders’ answers to studio audience members’ questions.

On three of the five measures, the greatest number of viewers ‘tended to agree’ that the leaders had spoken satisfactorily. Only minorities of a tenth or less ‘strongly agreed’ on the one hand or ‘strongly disagreed’ on the other.

But for supporting their claims and arguments with factual evidence and for ‘understanding people like me’, majorities of viewers tended to disagree or disagreed strongly with the statements concerned – in the latter case totalling 59 percent of
the respondents.

The figures in Table III, showing how the Question Time viewers rated Jeremy Corbyn and Theresa May individually in terms of entitlement delivery, are telling in this context. On all five measures, Corbyn was rated more highly than May – with the margin of his lead being greatest for ‘understanding people like me’.

Young voters and the debates

The 2017 General Election campaign saw an upsurge in the proportion of young electors who turned out to vote on Polling Day. Much has been written and said about the political sources of their greater determination to cast ballots this time (such as issues of tuition fees, housing shortages and prices, poor job prospects, etc.).

But apart from the role of social media campaigning and doorstep canvassing, little has been said about the communication factors responsible for their heightened involvement in British politics in 2017. Our research may shed some light on their roles in that development.

Interestingly, this part of the story begins some years before 2017. Over the post-war period, participation in politics and exposure to political information tended to increase steadily with age in many democratic countries, including the UK. But signs of a more complex and changing role of age in political and communication behaviour emerged from the findings of our first study of voters’ orientations to the three televised Prime Ministerial debates of the 2010 General Election campaign.   

Although young people were still less likely than older citizens to vote at that election, some of their perspectives on and reactions to the three prime ministerial debates in the preceding campaign were either very like those of older voters or on some points even more positive. Rates of debate viewing and willingness to stay tuned to them to the very end were similar across all the age groups.

18-24 year-olds were more likely to feel that they had got things out of watching the debates than had older voters – ‘learning something new’ from them, for example, learning more about the policies the parties were proposing, and (in many cases) both looking forward to the debates in order ‘to help make up my mind how to vote’ and declaring after Polling Day that the debates had indeed helped them to make up their minds how to vote.

From such evidence, we concluded that ‘by and large, the youngest voters… seemed almost to have formed a special relationship with the prime ministerial debates’ (Coleman, 2011, p. 43).

We carried out follow-up research on the reception of TV debates in the 2015 and 2017 campaigns, replicating some of our 2010 measures while supplementing them with the measures we had devised of entitlement expectations and evaluations.

In the 2015 study, findings from measures of the latter responses showed that over and over, with few exceptions, younger voters’ assessments of the likely and actual delivery of the five entitlements were more positive than were those of older voters, on some points by quite sizeable margins.

For example, 19 percent more 18-24 year-olds than those aged 55 and older were confident that the participants in the first 2015 debate would ‘prove they understand people like me’. And after seeing each of the three debates transmitted during the 2015 campaign, young voters’ assessments were more positive on at least three of the entitlement criteria.   

Partly because they are most comparable to the focus of our 2017 study, we illustrate this point by showing the sample’s reactions to 2015’s ITV televised debate, featuring seven party leaders, in the following table:

Table IV

Percentages of 18-24 year-old and 55+ age groups agreeing strongly or tending to agree that in the 2015 ITV debate the leaders

The findings of our 2017 research portray a decided leap forward in younger voters’ levels of campaign involvement. Age stood out as a particularly discriminating variable (for young people especially) in 2017 than in 2010 and 2015.

To particularise: this election, members of younger age groups were almost as interested in politics as older voters were. The differences between them had narrowed significantly.   

In 2015, 50 percent of the 18-24 year-olds described themselves as at least ‘fairly interested’ in politics, compared with 75 percent of voters aged 65 or older. In 2017, the relations concerned had to some degree been levelled up, with 80 percent of the 18-24 year-olds, 75 percent aged 25-34, 72 percent aged 35-44, 78 percent aged 45-54, 79 percent aged 55-64, and 80 percent aged 65+, being so inclined.   

More young viewers watched the entirety of the Question Time Special than older voters – 56 percent of the 18-34 year-olds compared with 48 percent of 35-54 year-olds and 46 percent of 55+ viewers. And more of them talked about the programme during or after its transmission – 93 percent among 18-34 year-olds, compared with 84 percent and 77 percent respectively in the other two age categories. This was especially true of their conversations with family members (43% so engaged compared with 37 percent and 30 percent of the older sample members) and with friends (33% compared with 21% and 15% of the older respondents).

Seeing the programme boosted young people’s interest in the rest of the campaign to an above-average extent – 52 percent of those aged 18-34 years declared that as a result of seeing the programme they had become ‘more interested’, in contrast to 41 percent and 30 percent in the older age groups.

Younger people claimed to have learned considerably more about British politics from the programme than did older ones. For example, 68 percent in the 18-34 year group said they now knew more about the policies of the main parties than before, compared with 51 percent of 35-54 year-olds and 36 percent of 55+ viewers.

Crucially, more of the younger electors said they had found the programme helpful in deciding both whether and how to vote on Polling Day. On whether to vote, the ‘endorsement proportions’ in the three age groups were 46 percent, 28 percent and 20 percent respectively. On how to vote, they were 45 percent, 34 percent and 26 percent.

This evidence is interestingly related to another age-based difference among our sample members. Although younger voters were more supportive of the Labour Party than were older ones, more of them were still uncertain about how to vote on Polling day – as many as 38 percent compared with 30 percent of the 35-54 year-olds and 17 percent of those aged 55 or older.

Relevant in this connection are the facts that 38 percent of those only ‘fairly certain’ about going to the polls said the programme had helped them decide to do so and that 37 percent claimed it had helped them decide how to vote.

These data favour the inferences that much of this uncertainty might have been resolved in the last week of the 2017 campaign and that seeing the Question Time Special may well have played some part in the process.

Younger intending- and actual- viewers of Question Time were also more inclined to feel that modes of address they were entitled to expect from political leaders would be and had been satisfied by participants in the programme.

Table V i

Very or Fairly Confident of the leaders’ Entitlement delivery:

This was especially the case for provision of factual evidence, engaging them in the debate, and ‘understanding people like me’, as the figures in the following table show:

Table V ii

Agreed Strongly or Tended to Agree that the Leaders Had:

Further support for the above-mentioned inferences comes from the answers that Question Time viewers gave when asked, ‘During the debate, which of the leaders, if any, was most effective in…?’ Interesting in this connection is the fact that assessments of Corbyn as a preferable speaker on all the entitlement measures were strongly associated with affirmations by the individuals concerned that watching the Question Time Special had helped them to decide whether to vote on Polling day.

As Table VI shows, younger people rated Jeremy Corbyn more highly than Theresa May for delivery of all of the entitlements – and especially for ‘understanding people like me’ and ‘engaging me in the debate’.  Amidst a tendency for the ratings of the two leaders to change steadily as people got older, first-time voters (followed by the 25-34 year-olds) were exceptionally impressed with Corbyn compared to May, while those of retirement age favoured May over Corbyn to an extent unmatched by any of the other age groups.

Table VI

Who was most effective for:

Other key factors

The forces that shape communication processes and the operative elements within those processes are multiple, inter-related and complex. Gaining a fuller understanding of the factors at work during the 2017 General Election campaign, beyond those we have already identified in previous pages of this report, remains a task ahead and there are three such influences that will undoubtedly merit further attention.

One factor is the role of gender which, though quantitatively less distinct than that of age, appears to have been significant.

At first glance, women on the panel declared a lower level of general political interest than men. Fewer of them had been actively seeking news of the campaign before Question Time was aired. Of those panel members who tuned in to it, fewer women than men watched the whole of it, while in the first wave of the survey women were less certain about how they would vote.

However, those women who did view the programme talked about it more than the men did, especially with partners and family members.  They also rated the leaders more positively than did male viewers for supporting their claims with factual evidence, engaging them in the debate and ‘proving they understood people like me’.

More women than men said they had become more interested in following the rest of the campaign, and claimed to have learned more about the policies that the parties were proposing. And they found Jeremy Corbyn to have been more effective in meeting their entitlement demands – most strikingly, considering his gender, for ‘understanding people like me’.

This points to a second variable that seems to have been critical in our analysis: the extent to which political leaders appeared to demonstrate that they ‘understand people like me’ was central to overcoming – or reinforcing – the tendency of voters to feel estranged from the political class.

The segments of the electorate who seemed most sensitive to politicians not understanding people like them, according to our data, were white, middle-aged and elderly males, more likely to reside in Scotland, Wales and Northern Ireland, and least likely to reside in London. They were more supportive of minor parties (Lib Dem, SNP and UKIP), lacking in general political interest, unimpressed with Theresa May’s effectiveness as a speaker compared to Jeremy Corbyn, and lacking trust in politicians’ ability to run the country well.

As we stressed in the introduction to this report, trust is a sine qua non of good representative government. As measured in our first-wave survey, over a third (35%) of respondents said that they had no trust in politicians to do what was right for the country. Several major factors seemed to have influenced attitudes on trust of politicians.

One was integration into the political system. For example, political trust was closely related to level of general political interest in our data. And as many as half of the undecided voters in our first-wave survey lacked trust compared with three tenths of Labour supporters and a fifth of Conservative Party supporters.

Another crucial source of influence on this sentiment was socio-economic status: the lower the social grade the least trust in politicians (41% no trust among DEs, 38% C2s, 34% C1s and only 20% ABs).

The relationship of age to political trust was strong but curvilinear – that is, individuals in the youngest and oldest age groups were the most trusting of politicians. Lack of trust increased with advancing age among 18-24, 25-34, 35-44 and 45-54 year-olds (29%, 30%, 40% and 45%, respectively) and then was lessened – 32 percent among 55-64 year-olds and 29 percent among the 65+ respondents.

Level of trust did seem to have affected political viewing behaviour – with 54 percent of those first-wave respondents who definitely did not intend to watch the Question Time Special having declared no trust in politicians. That compares with 39 percent of those who said they probably would not watch, 30 percent of those who would probably watch and 22 percent of the declared definite viewers of the programme.

Political trust also seemed to have affected people’s assessments of political leaders’ relation to communication entitlement.  For example, confidence that leaders would ‘prove they understood people like me’ when taking part in Question Time was 80 percent for those who trusted to a ‘great extent’, 50 percent trusting to ‘some’ extent, and just 15 percent among those with ‘no’ trust.

On the other hand, watching the Question Time Special did seem to have mitigated somewhat the apolitical consequences of mistrust. Although, for example, more trusting viewers of Question Time were inclined to say that it had fostered greater interest in following the rest of the campaign, as many as 35 percent of the untrusting viewers of the programme said the same. Over a quarter of those untrusting viewers said it had helped them to decide whether to vote on Polling Day. And nearly a third said that it had helped them to decide which party to vote for.

What this means for political communication and democracy

Firstly, we think that the notion of communicative entitlements should be recognised as central to democracy. Politics has tended to be been dominated by a supply-side imbalance, with more attention paid to politicians, their rivalries, how they play ‘the game’ and what they imagine the public wants than what citizens actually demand from a well-functioning democracy. The political class comes across too often as if engaged in a conversation with itself, neglecting the everyday experience of citizens.

The more we have come to think of democratic politics as a spectator sport, directed by communication strategists and filtered by professional pundits, the more citizen-spectators have come to doubt its relevance to them.

The findings presented in this report challenge all of us to re-think the democratic relationship, paying close attention to citizens’ communicative entitlements to be addressed, informed, engaged, recognised and empowered in ways that enable them to function as a mature demos.

Secondly, given that citizens benefit from media exposure to politicians who are given time to set out serious political arguments and are faced by meaningful, live challenge from sceptics and opponents, televised debates should not be regarded as an added extra within important democratic processes like election campaigns.

Debate is not only good for democracy, but a necessary condition. The Conservative line in 2017 was that Theresa May regarded televised debates as a distraction from “a traditional campaign where we can get out and speak to all the voters, so they see people personally” (Conservative spokesperson quoted in The Sun, 31 May, 2017).

But getting ‘out and speaking to all the voters’ often amounts to little more than stage-managed appearances from which media and public questioning is banned. A BMG Research survey commissioned by the Independent (25.4.17) found that 54 percent of the public thought that the leaders of the UK’s major political parties should participate in live televised debates during the 2017 election campaign, with more Conservative supporters in favour than against. And a Change.org petition calling upon the broadcasters to ‘empty chair’ any party leader refusing to take part in the debates attracted 121,966 signatories.

Once the idea of a head-to-head debate between the leaders most likely to become Prime Minister was vetoed, the media ‘debates’ that took place in May’s absence were strange events: a combination of non-debates in which opposition parties said what they would have liked to have said to Theresa May had she been there, and machismo interrogations by aggressive interviewers who seemed far too absorbed in their own polemical agendas to engage in anything resembling a reasonable conversation.

The Question Time Special, in which the two main party leaders appeared consecutively before a studio audience, was a valuable supplement to head-to-head debate, but not a satisfactory alternative. In the next election voters deserve to have an opportunity to watch both forms of televised debate.

Thirdly, we have already suggested the considerable importance that citizens attach to the ‘understand people like me’ entitlement. On the face of it, this is an obvious requirement: why would people wish to be represented by someone who does not understand the kind of life they lead or problems they routinely face?

But to recognise the importance of this aspect, and to now have some empirical data to help illuminate it, is only to acknowledge a problem. Building democratic relationships that transcend the communicative insensibilities that have estranged people from ‘political elites’ calls for some fundamental changes to the ways in which we have come to think of political discourse.

Determining the terms of such changes and then deciding how to bring them about should itself be a matter of public debate – a debate to which we hope this report offers a constructive contribution.

References

• Coleman, Stephen, ed. Leaders in the Living Room: the Prime Ministerial debates of 2010: evidence, evaluation and some recommendations. Reuters Institute for the Study of Journalism, Department of Politics and International Relations, University of Oxford, 2011.
• Coleman, Stephen, and Giles Moss. “Rethinking election debates: What citizens are entitled to expect.” The International Journal of Press/Politics 21.1 (2016): 3-24.
• Couldry, Nick. Media, society, world: Social theory and digital media practice. Polity, 2012.
• Garnham, Nicholas. “Amartya Sen’s “capabilities” approach to the evaluation of welfare: Its application to communications.” Javnost-The Public 4.4 (1997): 25-34.
• Hesmondhalgh, David. “Capitalism and the media: moral economy, well-being and capabilities.” Media, Culture & Society 39.2 (2017): 202-218.
• Mansell, Robin. “From digital divides to digital entitlements in knowledge societies.” Current sociology 50.3 (2002): 407-426.
• Nussbaum, Martha C. Creating capabilities. Harvard University Press, 2011.
• Sen, Amartya. The idea of justice. Harvard University Press, 2011.

The authors would like to thank the Humanities Research Institute and the School of Media and Communication at the University of Leeds, together with the Electoral Reform Society, for funding this research. We are grateful to Giles Moss for his valuable contribution to our application of a theory of media entitlements. We appreciate the work of ComRes, especially Dan Holden.