Loading summary
Transcript26 lines
- [00:01]
A
Welcome to the New Books Network.
- [00:06]
B
Welcome to the People Power Politics podcast, brought to you by cedar, the Centre for Elections, Democracy, Accountability and Representation at the University of Birmingham. Hello and welcome. My name is Tim Horton. I'm a Professor of Comparative and European Politics at the University of Birmingham and a Deputy Director of cedar, and I'm going to be the host for this episode. It is my great pleasure to welcome Kate Dommett. Kate is professor of Digital Politics at the University of Sheffield. Welcome to the podcast, Kate.
- [00:41]
A
Great to be here. Thanks so much, Tim.
- [00:43]
B
Politics, parties and campaigning are all changing AI, digital tools and the rapid spread of messages all mean that the conduct and content of politics is changing. In many respects it feels like the only constant is change, but closer observation often illuminates a patchier picture with elements of change and elements that remain. Moreover, change can be more evolutionary than revolutionary, and the change is not always along the lines we might predict. So how and in what ways is political campaigning changing? What role are digital tools playing? What do citizens want from their political parties, and what are they or could be doing to meet those desires and expectations? To help us better understand campaigning, digital politics and public perceptions of party politics, I'm delighted to be joined by a terrific scholar who brings a wealth of knowledge to the table. Kate's research focuses primarily on digital technology and democratic politics, with a particular focus on data use, election campaigns and regulation. She is particularly interested in how politics is being changed by the uptake of digital technology and how democratic institutions are adapting to use offline affordances. Her research has covered political campaigning, political advertising, the regulation of digital politics, and how political parties are viewed by the public. Among her many publications are two books that form the springboard for a significant slice of our discussion today. Kate is the author of the Reimagined Democracy, Change and the Public, published by Manchester University Press in 2020, and a co authored book with Glenn Kefford and Simon Krasinski, Data Driven Campaigning and Political five Advanced Democracies Compared, which was published by Oxford University Press in 2024. A reminder that the back catalog of the People, Power and Politics podcast has a wealth of discussions, including the discussion I conducted with Maria Sobolevska, Charlotte Galpin and Monika Brusenbach Maislova on Brexit Britain 10 years after the referendum. We are recording this podcast on 22 May, following local elections and elections to the parliaments in Scotland and Wales that may prove to have significant consequences for the future of British politics. There is also lots of speculation linked to by elections Andy Burnham and Keir Starmer's leadership. By the time you are listening to this podcast, it is likely that the result of the recently called by election in Makerfield will be known, whatever the outcome of that electoral test is. I, however, would venture to suggest that the churn and change of British politics is likely to continue. So I'd like to begin by actually looking back to 2024. The last UK general election campaign was memorable. From a sodden Sunak announcing the election on the steps of Downing street to end Ted Davies succession of stunts, there were indeed many memorable moments. But in the months preceding that election there was much talk, in particular in journalistic circles about the role of AI, that this could be the kind of first AI election. Now, in your chapter in the recently published book the British general election of 2024, you actually write, however, that despite much speculation about more sophisticated and AI informed campaigning, the 2024 digital campaign was in many ways simple and routine. So I'd like to ask you, what role exactly did the digital play in parties campaigns in that election, and how were those digital tools actually employed?
- [04:33]
A
Thanks. It's a really great question to get us going, so I think it's probably useful to start by saying it's a good kind of rule of thumb to be a little bit cautious of the media narratives around digital technology. Because every election cycle there is always a clamor to kind of proclaim the new election. And this time around it was the AI election. Previous times it's been the Facebook election. You might also recall the TikTok election. Journalists are always looking to really profile what the latest development is and to focus on that as the news story. But I think what my research has consistently tried to do is actually, actually look at the practice of what is actually going on. Are, you know, concerns that are being raised in the popular media, valid or not? And I think what we can say if we think about 2014, is that digital is a really important part of campaigns, but it's not yet the main component. So I find it hard to imagine a campaign in the UK that would be conducted entirely using digital technology. So it's really important that we kind of understand how digital is being enmeshed alongside those more familiar techniques such as door knocking, leafitting hustings, and kind of political speeches and just traditional news coverage. Now I think digital has become increasingly important and that's because some of those more traditional activities of campaigning are quite hard for political parties to do nowadays. So they have less members, they have less activists, people are also just less willing to open their door to strangers and talk about politics. So there are a lot of reasons why digital has become an important component. And that is kind of something that we saw in 2024. It is undoubtedly a really important way that campaigners are trying to connect with voters and get their message across, either to kind of the entire population or to particular segments of that population. And I think what's particularly interesting thinking about 2024 compared to 2019 is that we're kind of having to see political partisan campaign evolve in response to changes in the online space. So if you think back to 2019, that election was really focused on Facebook and there was a lot of kind of coverage of how political parties were active on Facebook and that reflected where we were online. Now roll forward to 2024, and there's some really lovely data from Ofcom about where people are online and what platforms they're using. That's showing that there's been a fragmentation of where people are online. And what we've seen in response at 2024 is political campaigns try and replicate that and be active in more spaces than they were previously. So this is why we saw the kind of creation of TikTok's accounts. This is why we've seen political parties using WhatsApp, using YouTube, pre rolls, for example, is that that's where we are online. So it's a kind of constantly adapting process, but at the core of it, it's about political parties using any mechanism possible to actually try and reach out and connect with the electorate.
- [07:43]
B
Yeah, that's interesting that ultimately we can say a lot about campaigning, but it's ultimately about communication and. And about the effective means of communication. And parties use whatever tools are available to do that. I think you've given us a really good sense there of kind of what was being used. I guess the next question I would like to throw at you is sort of reflecting Both on the 2024 British general election and indeed the more recent elections in the uk, local elections, recent by elections, elections to the Scottish Parliament and the Welsh Senate, for example. Do we have a better sense of what really works and doesn't work in campaigning, thinking about the use of these tools, and perhaps we might also ask, how far have we seen parties learning from previous experience about what works and what doesn't work that's actually informed their practice when it comes to campaigning?
- [08:43]
A
Yeah, this is probably one of the hardest questions you could ask someone interested in campaigning, because studying the kind of effect of actually what works, I think is really hard. And there's different end goals that you could want here. So are we talking about what works in persuading someone to vote a certain way, what works in mobilization, or just what works in getting attention? So they are kind of very complex questions and they're very hard to analyze. But you know, the prevailing research shows that the most effective form of campaigning is still kind of door to door, door knocking, face to face conversations, and that real, very offline personal contact, both in terms of mobilization and then to a lesser extent around persuasion. Believe it or not, we're quite hard to persuade in our political views. So what's interesting is that online effects are relatively small. They are present and it's very difficult to effectively test what seeing a piece of content on Facebook or YouTube has on you. Because actually what we think is likely to be happening is it's a kind of multiple exposure effect. You see so much content online that it's quite hard to measure kind of the impact of an individual thing. So I think there's kind of that question and I think honestly that is a large motivation of why we've seen campaigns not forsake those traditional mechanisms in favor of an entirely digital campaign. But I think there's a kind of a different kind of question as well, which is what the parties and campaigns feel they have to do. And I think what's been really interesting is that there is this feeling, and I think it's been manifest in different election campaigns that you need to be active in certain places in order not to risk kind of being seen to fall behind your competitors. And I think that's really, for me, the story of TikTok. So, you know, TikTok we know is kind of used by a particular demographic, by and large, younger voters. We know that for thinking back to 2014, it wasn't really those voters that a lot of parties were going after and trying to secure a majority, but nevertheless, they felt they had to be on those platforms because the media coverage kind of turned into, you know, who is on TikTok? What are they saying? And we saw throughout the election campaign parties join in turn and make sure that they were part of that conversation. And I think that's quite interesting in that there's clearly almost like a practice of campaigning where parties feel that they need to be active in certain spaces. And so it's almost leading to a convergence around what parties are doing. I think the only kind of exception to that is that you do still see differences and they're explained by kind of quite mundane things. Which is you do parties literally have the people, the expertise and the money to be doing all of these things? And they really don't. And I think that's kind of where we see smaller parties tend to focus on one or two platforms and not really have a lot of content going on online, whereas we see parties with bigger budgets really kind of experimenting and having more paid content online particularly.
- [11:54]
B
Yeah, that's a fascinating insight there. And actually I'd like to broaden out from the kind of UK case. So looking beyond the United Kingdom and placing its experience in comparative perspective, I'd like to turn to the co authored book that I mentioned in my intro, which looked at data driven campaigning. So just for the benefit of listeners, can you give us a definition of what you understand data driven campaigning to actually mean?
- [12:22]
A
Yeah, so. So I'll give you two routes into this. So when someone asks me in a taxi what I do, I say I study whether what was covered in the Cambridge Analytica scandal is actually going on. And I think that tends to be the most kind of pertinent entry point for people is it's, you know, this idea that political campaigns were harvesting large amounts of personal data about us, often kind of very intimate information about our sexual preferences, our marital status, you know, our interests and attitudes. And we're basically exploiting that to be able to try and change people's views without their knowledge. So that's the kind of popular explanation. But I've tried to define this in kind of academic context in two places. So firstly, kind of in the book, it's defined as a mode of campaigning that seeks to use data to develop and deliver campaign interventions with the goal of producing behavioral or attitudinal change in demographic systems. And then we kind of look at four components of data driven campaigning, which is data analytics, technology and personnel. I've got a slightly more kind of techy definition as well, which is in an article in Information Communication in Society. And this one was based on a systematic analysis of all of the kind of existing literature that has been published on this topic, stretching right back to over a decade. And that kind of really focuses on how data is used to inform decision making in either a formative or an evaluative capacity. And it's employed to engage in campaigning efforts either around voter communication, resource generation or and internal organization. So a little bit more techy there, but I think that second one kind of really gets into the way that data can be used in multiple different ways. It isn't just simply about making decisions, it's also about understanding if what you're doing is working.
- [14:17]
B
I find, I mean I really enjoyed the book and I found, I mean this, this notion of kind of getting us to understand how data is used to also, you know, particularly to kind of enhance traditional means of campaigning which comes out really well in the recent local elections a friend of mine was running for re election and I went out to campaign with him and I found it quite interesting to see the way in which the information that had clearly been put into the app was then used. And then they said right Tim, can you go to number 23 to knock on Kate's door? She's wavering a little bit but could be persuaded. Or it would say, you know, go over to number 53 and Steve is a postal voter and the postal votes will be out already. So you just need to firstly find out whether he's voted or not. And I found that very interesting that ultimately, as you rightly say, much of campaigning is actually old fashioned but it's the way in which data and the harvesting of data can be used in ways to make it kind of more effective. As you've mentioned, the book, we'll call it ddc so data Driven Campaigning to the audience. So the book covers kind of DDC in a number of countries. You cover Australia, Canada, Germany and the US as well as the United Kingdom. So particularly of interest to someone like me as a comparativist, I'm interested to know what did you find in your research about the use of DDC and how did this vary across these five countries that you, that you examined with your co authors?
- [15:53]
A
Yeah, so this was a kind of project that was motivated by a bit of a frustration about the dominance of work on the US So actually in the book the kind of US acts as a little bit of a comparator case and we really focused on showcasing what was going on in those other four countries because there is a kind of tendency in the study of campaigns to look at the US because it has historically been so influential in shaping campaign practice elsewhere. But we were kind of seeing on the ground in the countries we were based in that it wasn't quite playing out the way that we were seeing in the US and one of the motivations behind the book was firstly to kind of really actually just lay out what is going on and to systematize what these terms meant. What does data mean? Actually there are a whole range of different kinds of data and then to have a look and be like well what kinds of data are showing up where and why might that be? So the book has kind of two aims, really. It's quite. It's about offering kind of categorizations and conceptualizations of each of these things and then trying to explain those variations. So we looked at five countries, but we also looked at 18 political parties within those countries. And I say that because it's a lot of the variation that we detailed was actually within country as well as between country. And I think that's really important because it leads us to look to a set of kind of party level explanatory reasons as to why data was being used in different ways. So let's kind of just take one example. So if we think about data, one of the things we do is kind of talk through the different types of data. So you know, you can have state provided data, so census data, or we have the electoral role. In the UK you can purchase data. You can do so either through kind of polling and gathering information that way, or in some countries you can buy data sets from commercial providers and integrate them into your party data set. In other places, you, you know, there's histories of doorstep campaigning and that kind of data collection that you were just talking about. But as you can kind of already see in how I'm framing this, different places have different cultures and norms and systems of regulation, which actually mean that each of these sources doesn't look the same in different places. So the electoral role is probably one of the most essential sources of data in the uk. We've got some really nice interview data from campaigners across the political spectrum saying that is the bedrock of every data set within the UK is knowing where someone lives. Because geography is so important to our political system. Knowing if they've got a postal vote, as you mentioned, and knowing if they turned out and voted last time. Because we also have the marked register in the uk. Now, the US does also have a kind of data set of electors, but that varies by state. It contains very different information. If you look to other countries, there isn't an equivalent. So they're starting from a very, very different start point to where a campaigner in the UK or the US is. Now, in other countries, you may be able to purchase those data sets, but you can kind of start to see that actually quite quickly you end up with a very, very different set of data. Now, one of the things we kind of talk about is those four elements of campaigning that we differentiate. So data, analytical tools, technology and personnel, each of those four things varies in different ways. And we work to categorize Them, and fundamental to our kind of argument is that there are three explanatory factors that you need to pay attention to. So there's the party level, the kind of regulatory system and then kind of systemic factors. So at the systemic level, that's things like electoral system. So the fact that there's compulsory voting in Australia means that you don't need to kind of do the analysis to mobilise people that you need to do in the UK regulation. So this is like, which data sets can you use? You know, can you work with an external campaign organization? Or are there rules against coordinated campaigning, for example, that will affect whether you're able to share data? But I think one of the things I found most interesting was the party level. So we've already spoken about resources, but attitudes to data turned out to be a really important thing. So if you have someone at the top of a political party campaign who has really close ties with a data analytics firm, which was the case in one of the political parties in Canada, then data analysis becomes quite central to the way that the campaign organization runs. But we heard of other parties, and particularly kind of green parties, that were very, very skeptical about data and had real concerns about privacy. And then you saw not only a reluctance to adopt those tools, but, you know, the kind of politicians and the activists on the ground often, you know, would kind of sabotage these things because they didn't feel that it was the correct way to be going about doing politics. So essentially there's huge levels of variation, but I think those factors really help us understand and be able to predict a little bit more why we're seeing what we're seeing in these different places.
- [21:07]
B
That's really interesting. I mean, a couple of things that struck me about the book, which I liked. I mean, firstly, I very much agree with the sort of. The problem with a lot of the campaigning literature that it tends to be very US focused and that seems to be often the kind of benchmark against which we measure things. But in so many respects, the US is exceptional as important it is to look at campaigning in the United States, I think that those of us who look at other parts of the world would sometimes be better off not trying to take theories directly from that, but look elsewhere. Secondly, I mean, one thing I wanted to commend about the book, which I enjoyed, is that you blend together the quant and the qual, you conduct quite a lot of interviews, which I think, for an awful lot of work on campaigning, would think, well, how does this fit into a nice, neat data set? How am I going to create a regression test based on this? But actually that was very illuminating and to bring out some of the differences. I mean, one of the, I think I remember correctly because you talked about regulation. One of the things that interests me is that I think it was in Germany that you report that some data guys came to talk to a political party and gave them 11 recommendations. And then one person in the audience said, but seven of those are against the law. So you know, there might be great ideas for campaigning, but we just can't do it because of the legal restrictions. I mean, I thought that was, was really interesting. And the final point that you just mentioned just now about the data and the attitudes to the data, I think is really interesting. And of course we're in an environment now where all the time we're also giving our data every time we go to the supermarket or wherever. We seem to be handing over so much data. And of course data driven campaigning is contingent very much on data. So the next question I really want to ask you about is given that this kind of key ingredient of DDC is sort of good high quality data and you know, as we've touched on one of the really striking findings in your book, stress that parties ability to collect, access and integrate data is far less extensive or sophisticated than frequently depicted. Now you and your co authors are sagacious enough to recognize the velocity of changes in the technological world and I might add in here of course the slow pace of academic publishing, which is a bit frustrating when you write about these kind of things. So I wanted to ask you from your more recent research and observations, I mean, how far would you actually say that the quality of the data collected has improved and the processing of those data has become more sophisticated and more effective in more recent elections or is still the quality of data may be not that high and hence may in fact explain the limitations of ddc.
- [24:01]
A
Yeah, so this is kind of one of the things that actually motivated my, my interest in this topic. So Rasmus Nielsen wrote an absolutely brilliant book called Ground wars, which was an ethnographic account where he embedded himself into campaigns in the US And I remember there's this one bit where he talks about observing people do data collection and he kind of talks about what was going on. He's saying, you know, sometimes you'd get a really kind of tired activist and they would just cut corners and just enter some data really quickly because they just clearly really wanted to go home. You know, they might put, you know, they're not in when they knocked the door instead of waiting. And it's something that I've kind of observed in following various campaigns in the UK myself, is that, you know, it's amazing how that kind of every day, you know, supposedly the core of so many campaigns, data collection, that ground campaign, is based on really, really shoddy data. So, you know, I've seen myself, people stand on doorsteps, talk to one person, walk away and go, oh, brilliant, five labor voters in that household, or oh, yeah, all three Greens there, this is great, they'll definitely turn out. And that's a massive assumption because why would everyone in a household vote the same way? So I think those kind of anecdotes for me start to illustrate that a lot of this data is quite unreliable, particularly if it's kind of gained through activists who may be trained with varying degrees of rigor. You know, these often people are giving up their time and who are interested in having conversations. They're not automated data gatherers. And even kind of other sources that, you know, we think of as more rigorous, like, you know, polling or focus groups, you know, as academics, we know that both of those mechanisms have their flaws and challenges in getting accurate and kind of truly representative insights. And what that means is that basically across the span of data sets that political parties and campaigns hold, there are flaws and inaccuracies within the data that they have. There are also huge gaps. But I think one of the real challenges that politics face is this events. You know, in an era where we've had quite a few prime ministers and a lot of political changes, that data goes out of date very, very fast. And if you're reliant in the main on activists to go and collect it for you, it becomes a real challenge to actually assume that your data is up to date. And there's some really nice accounts of the 2019 election, which is kind of saying that that's where the Conservatives really fell down, is that they assumed that the data that they'd collected previously would be, would put them in a good place. And actually it was very out of date because attitudes have changed so much since 2015. So I think that that means that we really need to kind of pay attention to data and the kind of fundamentals. And so I've kind of been thinking about this in the most recent piece I've had published, which is kind of thinking about the potential for automation in data driven campaigning, which has just been published in Communication Theory, because kind of angling behind your question, I think, is there has been a lot of change, you know, the rapid evolution in AI that we've seen over the last year has really shown the sophistication of now what it's possible for people who previously lacked expertise to be able to do in the online space. And that theoretically opens up a huge new toolkit for political parties in campaigns. But it's premised on the assumption that you've got high quality data. And you know, there's some really nice literature around automations. There's an article by Davenport and Harris I used for this piece that's kind of talking about, you know, you can automate decisions if you can readily codify decision rules and have high quality data available. And then those decisions need to be kind of made frequently and rapidly so that those are the kind of decisions that are really amenable to automation. And if you look about what we know about ddc, as I kind of do in this piece, you can see that most of those criteria just can't be met. It's quite hard to come up with fixed and rigid rules for data that can range from focus group to canvassing to polling data. We know it's not high quality as we've just discussed, and the decisions aren't being made with enough frequency and regularity to make it amenable to automation. So I think that whilst there are some spaces in which we might, we can say yes Maud, technical, sophisticated things could be happening. There's also a lot of reasons why actually much of the decision making is going to be quite mundane and quite traditional as opposed to automated and very data driven.
- [28:43]
B
That's fascinating. So far we've basically focused on your research into campaigning in the digital, but you've also conducted research which examined citizens views of political parties. In your 2020 book, based on research in the UK, you argue that there is this widespread desire for a kind of expansive reimagination of the way that parties operate. Close quotes. You argue that people want existing aspects of party politics such as aggregation and mediation, responsive and responsible governance and partisan leadership to be performed, but in a. In a slightly different way. So what I'd like to ask you first though is just to ask you, how did you go about undertaking that research for that particular book and why did you employ those particular methodological tools?
- [29:41]
A
Yeah, thanks. This book was so much fun. I think as a young scholar I was really interested in that kind of body of work about the decline of political parties. And you know, we have so many seminal accounts. Peter Meyer's work was a real kind of influence on me. Looking at, you know, varied Accounts of declining partner membership, partisan affiliation waning, and this idea that essentially parties were probably going to scrap heap. And I was really kind of interested in exploring why, despite this decline, parties are still very much a part of our system and they remain kind of essential linkage organizations. And I felt that that kind of meant that maybe we needed a slightly different approach to understand what was going on here. You know, maybe we'd actually needed to go and talk to people about what it was they wanted from these political organizations, how they felt they were performing, to understand a little bit more about why parties had kind of fallen into this set of negative perceptions, I suppose, and why they weren't seen to be performing in the way that perhaps once they had done. Although I'm a bit of a skeptic of the old golden age of parties. So the book is essentially really about collecting data from people. And I was particularly interested in expectations. So how did kind of people think that parties were performing currently on a range of indicators of democratic representation and linkage, and how would they ideally want them to perform? So to do that, I did two things. So the first was a nationally representative survey to kind of gather data on a range of survey questions. But the thing that was a bit more experimental and was absolutely great fun, if slightly mad to do at the time was I did these kind of what I ended up calling deliberative focus groups. So the goal was to have kind of 75 members of the public, and we had three groups, so 25 people in each group. One was the party activists and party members, the second was just the general public. And then the third was a mix of party activists and members and the general public. And essentially what we did is they were four hour long sessions and we kind of brought people together, sat them all round tables, got them to do a throat clearing, you know, what's wrong with political parties? What do you think of about them? But. But then we kind of had a series of exercises, a little bit of education, a little bit of reflection, to really help people think in a little bit of a more detailed way to kind of talk, to learn and to reason through kind of complex challenges around democratic representation to get a little bit more at how people wanted parties to change. And so then I combined those two methods. So throughout the book, the quantitative and qualitative data is presented alongside one another. So we look at those survey indicators and then I try and understand what might be explaining this. How did those comments in those four hour groups maybe help us understand exactly what people think to be wrong? And exactly what it is that they are after in the future.
- [32:52]
B
Yeah, I mean, I have to say it's one of the things I enjoyed a lot about the book. That kind of, that methodological approach, the blending together. I mean, in my own work I've used a lot of surveys which are great, but most of the time you kind of want to follow it up. And having the opportunity to bring people together within a focus group and having a kind of genuine conversation, I think is really quite fascinating. I mean, shock horror. If you want to understand what voters and citizens want from political parties, you have to talk to them and listen to what they have to say. But I think, yeah, the way that you used the focus groups in particular, I think were really, really interesting and generate some nice enjoyable quotes and observations in that. I remember the one person saying to you something like political manifestos are like kind of catalogues in travel agencies. You know, all looks great, but then you go on holiday and it rains for two weeks. Solid kind of notion. And I thought it brings out the color, I think really quite well. Now you've touched a little bit on the findings as well. But let me ask you a bit more specifically about the findings. So, I mean, one of the aspects of your book that chimed a real chord with me, especially in light of my own research on party politics in Central and Eastern Europe, is about democratic linkage between citizens and those who govern over them. You flag up three important elements of that linkage. Representation, participation and governance. If we take the first of those, representation, you draw on debates often associated with Edmund Burke seeing different styles of representation, as a trustee, as a partisan and as a delegate. So I wanted to ask you to sort of give us a flavor of what you found in your research about what kind of representation citizens in the UK actually want.
- [34:52]
A
Yeah. Goodness, what do people want? Always a fun question to ask. So, yeah, the kind of use of the trustee delegate partisan, you know, this is incredibly established. But the reason I was quite interested to build on this was Chris Carman had a really nice article that had tried to explore some of these issues and I was really kind of keen to look at the. The expectations gap idea around this because I felt like that was a new approach to bring in. And throughout the book we use Sankey diagrams, which if your listeners aren't familiar with these wonderful diagrams, where you can kind of trace the variation between two similar answers. So you can kind of see maybe 30% of your respondents have the same answer between two subtly different questions. Some of them might go up to A different question, some of them might go down to a different option. But it's quite an interesting way to allow us to kind of of understand the patterns and how people respond to similar questions. So essentially what we did was, so I asked this question around what kind of representation would you like to see which were. And we had little phrasings to kind of make trusty delegate and partisan a little bit more intelligible to your average person. But what was really interesting is kind of when you ask people what they like, you know, there's actually support for pretty similar levels of support for each of these different mechanisms. So kind of trustee and delegate representation was favoured by 72% in each case, whereas partisan was 66%. So we weren't asking people to compare, it was kind of with it, you know, do you strongly agree, agree or disagree? Strongly disagree. So there was kind of. I think that's an interesting finding because it shows that kind of each representative style actually has appeal. And I was quite, quite surprised that partisan representation featured as approved of as compared to the others. I would have thought that delegate, you know, that idea that you're acting on behalf of your constituents would have been a little bit higher. I think what was interesting is that the kind of explanation probably came out in the workshops because although we did hear people who really, really dislike dogmatic partisan thinking, there was consistently this emphasis on the importance of value led decision making and people really wanted their representatives to be principled and to have ideals and they felt that that was a really important trait. So I think that probably helps us understand maybe that surprising finding. But what was quite interesting is then when you kind of turn to how people felt that parties were actually behaving, we found that there wasn't kind of much difference between partisan behavior. So 66% said they thought that was great and they felt that 61% of people felt that that's actually how parties were behaving was in a partisan way. But then if you look at tristy and delegates, so it was 72% felt that these were good ways to represent in both of those instances. But then when thinking about parties, actually only 42% felt that parties were behaving as trustees and only 36% felt that they were behaving as delegates. So I think that kind of idea that parties were acting in the kind of common good and you know, taking that responsible mode in that kind of trustee sense or that they were acting on the interest of their constituents really wasn't being found in practice.
- [38:14]
B
Yeah, I think that's really interesting to see that gap now. You've given us a really good sense of the kind of general view of the role that parties should play. But I guess one question I'd like to pose in an era of very sharp divides, what political scientists often call about this kind of effective polarization of a lot of politics these days, does the story kind of change when citizens are actually asked about the party that they tend to vote for, for or have a strong partisan attachment to? Or is the story kind of the same so they want just the same amount of value led decision making whether they're talking about the party they vote for or parties in general?
- [38:59]
A
Yeah, so I think a little bit of both actually and one of these is quite surprising. So I think throughout my work, you know, anyone who has looked at a quantitative data set around this is kind of saying partisan bias is increasing incredibly strong. You know, people consistently talk about and give more favourable views for the party that they support. But I think one of the most interesting things is kind of we looked at this variance between how people thought parties should and were behaving. And what was quite interesting is that it didn't make a difference which party you kind of thought about into how you were answering those questions. And I think that's a really interesting finding for parties. Now making recommendations in this book was really hard because some of the findings were mild contradictory because the public never like to give you nice and simple solutions. But I think what I would find reassuring as a, as a kind of party leader or as someone in HQ who was potentially reading this book and thinking maybe they need to think about how they're behaving is that it was people across the spectrum were voicing this gap between expectations and ideals. So it's not as if they're responding to an outlier group or a group with a particular, particular party affiliation. There was this actual general feeling even amongst those people who were kind of activists, that a lot wasn't working correctly within political parties and that they'd like to see change.
- [40:23]
B
I really, I mean, we've mentioned Peter Mayer already. It's very difficult to talk about political parties without talking about the great scholarship of Peter Mayer. And one of the things I liked in the book is that you draw on his work and some of his work related to responsive and responsible governance. Again, this, I have to say, is something has very strong resonance with me, my own work looking at a different geographical part of the world now, thinking about the kind of responsive, responsible and representative demands of parties and how those demands Pull against each other. You've touched on this a little bit already about sort of what citizens actually want and whether parties could deliver that. Maybe. Maybe I should rephrase this question. Sort of say, you know, what could parties do? Do you think that would ensure they have a better chance of delivering what you found the citizens actually want?
- [41:15]
A
I think one of my kickoffs from this is there's a great book, a great quote from Dalton and Weldon in 2015 that described parties as unresponsive, untrustworthy, and unrepresentative. And I would say that a lot of the people that I spoke to agreed, but with that assessment, and so, you know, the easier answer might just be, you know, well, listen to the public more, maybe consider more kind of participatory modes of engagement or, you know, more opportunities for inclusive decision making. And I think there has been some interest in that. But I think actually, what the kind of findings I offer, it gives us a little bit of a different picture. So one of the kind of questions we asked is around level of party responsiveness. And so I kind of asked, do people want parties to be very responsive? Sometimes responsive, like to rarely change or never change, and kind of offer the different descriptors of adaptive, receptive, inert and immutable? Because we do love a term as political scientists. And I think what's really interesting is that when you kind of group those top two, so adaptive and receptive, so changing very often or sometimes. So 71% of people said they wanted parties to change very often or sometimes their kind of policy positions. But when you dig into it, only 21% wanted them to change very often. And I think that kind of symbolizes that there is a real tension here for parties. You know, we don't want. People are not looking for policymaking by referendum. They want parties to be open to listen and to respond to evidence, but they also want them to have some degree of consistency. It comes back to something I think we talked about earlier. So the. This idea of parties having values and ideas and consistent chords. I think that's a really useful idea. So, you know, people don't want parties to swing across the political spectrum and, you know, be very unique, unpredictable. But they do want them to take in evidence, to not have this dogmatic closed thinking where there is a lack of openness. So we did an exercise around whether people were interested in parties, like using experts as the source of policymaking. So we gave people a whole range of different people and processes that parties could use to inform their policy making. And experts kind of kept coming up again and again, but not in the technocratic sense. They didn't want experts to be in control, but they wanted to know that their decision makers were taking and listening to the best information that was out there and were reconciling that within the framework of a broad set of principles and values. And so I think that it's hard to kind of give a very clear recommendation about that, but I think it signals a certain ethos around parties. And I think that's what the book is really trying to get at, is we're not saying throw away the system and start again, but it's about thinking about the ethos of party democracy and how it is that parties are positioning themselves, how they explain the way that they're forming policies and ensuring that there is that set of kind of values and consistency that are guiding the way that they're approaching policy making. And I think that. That I would suggest is probably an approach which could be kind of enacted in how politicians actually explain what it is that they're doing. Hopefully actionable in some ways.
- [44:48]
B
I think so. And I think, you know, it's interesting. It's about, you know, what really shines through in the book is this need for, you know, on the one hand, kind of principled continuity, but on the other hand, an openness to kind of change. And it's really about sort of striking that kind of happy balance. And it's always going to be difficult. I mean, if we think about balance and we think about the weights on two sides, it's very difficult to get that sometimes. And it's a constant kind of task in order to try and see them in balance. But I think you provide, you know, your research provides really important and insightful ideas that could be employed or should be employed by. By political parties. But also I think it helps us also understand, I mean, it's something that you don't explicitly talk about in the book. From my memory of reading the book, you can tell me if you do say this explicitly. Apologies if you do. But in a sense, people like me who look at party systems that are very kind of fluid, that actually you provide some of the key ingredients, why we see fluidity, why we see, for example, in the part of the world that I focus on, lots and lots of kind of new parties bursting through and parties that seem to live fast and die young losing their support very rapidly. Now, we've touched on three key aspects of your published research. You've also mentioned this article on automation, which sounds fascinating. More recent work, an article I wasn't aware of until talking to you today. But I'd like to kind of close by asking you what you see as the kind of important and exciting questions to research in the field of parties and campaigning and digital politics and maybe also if you can give us a flavor of what you're currently working on.
- [46:41]
A
So I think I'll start at the kind of broadest level and I think I remain committed to that central tenet that we actually need to study what political parties are actually doing in practice and how they're changing on the ground. And that is increasingly hard. And I think poses are kind of real challenge for us as scholars of political campaigns and political parties, because certainly over the course of my career so far, it's much harder to get an interview with a party official than it ever has been in the past. So I don't think that's a kind of small undertaking. And I think it's something that we need to kind of stay committed to as a group of scholars because, you know, those kind of ethnographic on the ground really do help us work out whether these grand sweeping claims that we're making about party change and how campaigns operate are actually reflexive of everyday practice. But I think in terms of what I'm doing now, so I'm interested in two things, so one is slightly related. So I'm fascinated by the digital world. I absolutely love understanding what it is that that political parties and political campaigns are doing online. But it is incredibly difficult to study the online world. And so one of the things I've become very involved in is questions around data access and actually what information it's possible to extract about the online environment in order to enable us, and to be honest, regulators and people within government to actually be able to study and understand what is happening online. Because it's not easy to get an answer. And we are reliant on some quite imperfect tools to be able to gather information. So one of the things I'm working on now is with the social platforms Data Access Task Force is to actually lobby government to try and improve data access. And I'm really hoping this autumn there's going to be a consultation. I'm really hoping that we're going to see some policy change and that will really allow us to answer some very big questions about, well, how would we study the online if we actually had better data and we weren't just reliant on what there is? So I think there's a really interesting agenda there about how we how we study the online world, particularly in relation to elections. And then the other thing I'm working on at the minute is around regulation and how we effectively respond to the trends that we do know there. So I've gone down a lovely rabbit hole looking at transparency, disclosures and the information that we give to systems in political content and what effect they have, whether they work, whether political campaigners are implementing them effectively. Spoiler. No. And to actually understand, you know, in the context of we want our elections to be free and fair, we want people to have good information to be able to inform political choices. But at the moment, do we actually have good regulatory oversight interventions that allow that to be possible? I really don't think we do. And I think there is a real opportunity for change. And so I'm trying to kind of use my research to give us an evidence base to make some good regulatory interventions.
- [49:46]
B
That's really fascinating. I mean, I think you're absolutely right that we still, you know, we need to understand parties and we need to see what they're actually doing. I think the work on regulation, I think is increasingly important. It may feel like a rabbit hole, but I would just say it's a very important rabbit hole for someone to be going down. And the other point that you made, I think, which is crucial, which is both means it's important for us to continue research, but also provides excitement, is that it's very fast moving and we don't quite know what the digital sphere is going to be like in, in, in five, ten years time. I. We don't know quite how campaigning or political parties will kind of operate. But what is interesting is it's likely that we are going to again see significant elements which are continuity and certain elements of change. And the interesting thing will be to, to try and work out what continues, what changes and what why those two elements are part of the process. So I think lots for us to think about and to research in the coming years. So I just want to close by saying two words of thanks. So firstly, thank you to you, Kate, for coming onto the podcast.
- [51:07]
A
Thank you so much. I really enjoyed it and great to be part of an excellent podcast series. Thank you very much.
- [51:13]
B
And I'd like to thank you, the listeners, for joining us for this conversation. We hope you enjoyed it and I would just also encourage you to follow the links in the show notes to the two books of Kate that I mentioned in my introduction. Thank you for listening to the People Power Politics podcast brought to you by Cedar, the Centre for Elections, Democracy, Accountability and representation at the University of Birmingham. To learn more about our centre and the exciting work we do on these issues around the world, please follow us on Twitter edarbham and visit our website using the link in the podcast description.