Examination of witnesses

Gideon Skinner and Dr Glenn Gottfried.

Q50 The Chair: Good morning, everyone. Good morning, gentlemen. This morning, the committee is taking evidence on its inquiry into the voter ID requirement in elections. We are going to hear from Glenn Gottfried, an associate director in public affairs at Ipsos, and Gideon Skinner, the head of politics research in public affairs at Ipsos. I appreciate that your responses to our questions will be broadly within the parameters of your research—we will not take you into areas outside those parameters—but we will look at ways in which you can assist the committee in being more informed about the experience of the population, either by looking at the experiences of those who have experience of voter ID or by looking at what we can extrapolate from your survey in terms of what they might experience going forward.

I do not know whether you want to make any opening comments. We have a lot of questions; I hope that you have had an indication of what might come up.

Gideon Skinner: We are happy just to take questions.

Q51 The Chair: Okay; that is fine. I will open with the first question. To what extent did awareness of the voter ID requirement improve between the first and second waves of your survey? If I have got the timing right, the second wave took place after the local elections; the first wave was in January while the second was in May, after the elections. Can you tell us what they revealed?

Gideon Skinner: Yes. There was a clear sign that awareness of the need to provide photo ID increased over that period. If we look at our figures, awareness—at this stage, I am talking about photo ID as a whole before we get into VACs—increased from 34% to 76% between January and May. That looks like clear evidence of improving awareness. Awareness was good among all groups but it was lower among certain groups. We saw slightly lower levels of awareness among young people and ethnic minorities; these groups will continue to crop up in terms of where we see differences. Most obviously, awareness was particularly low among some groups that are less interested in and engaged with politics. Nevertheless, there was a clear sign that awareness increased considerably.

The Chair: I think that some of my colleagues will interrogate more on the demographic breakdowns but, on that evidence of improvement, was there a significant difference, in terms of participating in the survey, between groups from geographical areas that experienced local elections and groups from areas that did not? Could you give us a feel on that?

Gideon Skinner: Yes, we can.

Dr Glenn Gottfried: Please give me a moment. Fortunately, I have brought the datasets with me. From what I recall, the results showed that awareness increased among those who needed voter identification or lived in areas where it was required—I am sorry this is taking so long; it is a large dataset.

The Chair: We can come back to it when you find it.

Dr Glenn Gottfried: Okay. Gideon can take the next question while I dig it out.

Gideon Skinner: Let us carry on.

Q52 The Chair: Your survey suggested that awareness of the availability of voter authority certificates stood at 21% in May 2023 after the elections, whereas the Electoral Commission’s research places awareness at 57%. Can you explain the difference between these numbers? We know that there is a difference, but why do you think it has occurred?

Gideon Skinner: We have looked for this. We tried to find the exact wording in the Electoral Commission data and where it came from. We have not been able to find it, unfortunately, so we cannot give a precise reason. It may be the result of it using a slightly different question wording or a different scale, for example; we know that that can make a difference.

Let me give an example. Our awareness figure is based on a four-point scale that goes from “a great deal” or “a fair amount” to “not very much” or “nothing at all”. The figure that we reported on as our measure of awareness used the top two boxes—that is, people who at least said “a fair amount”—because we feel that, if you want to target a communications campaign, you will also want to target those groups who do not know much about it even if they have heard something. However, if you do include the proportion of people who say that they have not heard much about it—at least not “nothing at all”—we get higher levels; we get about 46% saying that they have heard of it on that basis, which is closer to the Electoral Commission’s figure. Other than that, without being able to see the detailed way in the Electoral Commission got that figure, I cannot give you an exact reason, I am afraid.

The Chair: Yes. It is quite a big difference: 21% versus 57%. Are you still pursuing looking at that question?

Gideon Skinner: No. We have tried to look. As I understand it, we have tried to ask the Electoral Commission and contact it directly. We have also looked at the tables on its website and the reporting of that figure but we have not been able to find that specific question. We could find other questions around awareness but not that specific one. It is the sort of thing where, if its question was a simple yes/no question as opposed to one with a four-point scale, for example, it might give a slightly different answer.

The Chair: Okay. Having read the summary of the findings in your survey, I note that there is a sort of footnote under paragraph 2.5. It says, “Given the nature of survey research, it should be borne in mind that groups least likely to have a photo identification might be less likely to take part”. Can you give us a sense of the extent to which that might have understated the impact of certain demographics? Or is it not statistically significant?

Gideon Skinner: I suppose that we just wanted to make sure that we were being responsible in that by explaining the pros and cons of survey research and helping people to interpret such research. Even though we felt that this was quite a robust method of collecting the data, in terms of the methodology that we used, there is always a slight danger that the people least engaged in politics, for example, will be slightly less likely to take part in survey research.

At the same time, when we have looked at the proportions of people who have some of the main forms of ID, they look to be in line with some of the other data sources out there. We are reasonably confident in our data but it is always important to understand the limitations of survey research. Even if they are perfect, they always have margins of error and things like that, so we wanted to be clear about that. We know, for example, that it is usual to get people overclaiming how much they vote in elections; that is a standard response that you might get in survey research. We just wanted to make sure that we were being open about those. We do not think it makes a significant difference to the results.

Lord Foulkes of Cumnock: I ought to know this, but what actually was the question? Did you prompt the person being asked about voter ID? Was it included in the question? Did you notice anything different, or was your experience different? If you prompted them, you might get a substantially different response, would you not?

Dr Glenn Gottfried: I have the question right in front of me. The question was, “Before today, how much if anything had you heard about the proposal that in the future you will be required to show a form of photo identification to vote in person at a polling station?” The responses to that were “A great deal”, “A fair amount”, “Not very much”, “Nothing at all” or “Don’t know”.

Lord Foulkes of Cumnock: But it might have been a different response if you had said, “Do you notice anything different about voting today from the last time you voted?” without mentioning voter ID. You have really prompted them there, haven’t you?

Dr Glenn Gottfried: Absolutely, yes. The difference between our figure and that of the Electoral Commission may very well be down to the way that question was worded, but we do not know how their question was worded so unfortunately we cannot comment on how much of a difference that might have made.

I have the figures now, if you were keen to know how the question related to individuals in areas that held local elections last May. Thirty-five per cent said they knew a great deal or a fair amount about the need for photo identification versus 63% who said they knew not very much or nothing at all. That compares to the figures in England as a whole of 31% and 67% respectively. That shows that there is a slightly higher number of individuals saying they know about the need for voter ID.

That was in wave 1. In the second wave following the election, as we mentioned, awareness increased, so 84% in election-held areas now say they know a great deal or a fair amount about the need to show voter identification versus 15% who say they know not very much or nothing at all. That compares to 81% and 17% respectively overall in England. So a slightly higher number of individuals are aware of photo identification in areas that held local elections last May.

Q53 Baroness Finn: Good morning, and thank you very much for coming. My question is about the voter authority certificate, which I notice was used by only 1% of voters in the May elections in any case. In May 2023 you found that just 25% of those without photo ID said they would be certain or likely to apply for a VAC. That has decreased from 36% in January 2023. Do you know the reasons behind that decrease? Did you ask?

Gideon Skinner: We did not ask a follow-up question to that so we cannot give a precise answer. There are probably two points that we can make. First, the proportion of people who do not have photo ID is a relatively small one so we would expect a correspondingly larger margin of error, given the number.

The other point, which could explain it, is based on our experience of doing research into elections in the past. We often see higher levels of certainty to vote in the run-up to an election as opposed to after one. It may be that in January, before the elections had happened, there was a greater sense of urgency to get a VAC, but then in the post survey, after the elections had happened, that sense of urgency or relevance had perhaps dipped slightly. It is possible that it was something to do with that, but we did not have a precise question on it.

Lord Anderson of Ipswich: You said the second-wave questions were asked after the elections in May, so the question “Would you be certain or likely to apply for a voter authority certificate?” might seem a little odd. Did it mean effectively “Did you apply?” or did it mean “Would you apply in the future at some unspecified date?” What did people understand by that question?

Gideon Skinner: It was a forward-facing question. It was similar to the question that we asked in the January wave because we wanted to be able to track and compare change. We asked in another question whether people had a voter authority certificate as one of the list of potential forms of ID that people could use, and the result was pretty low; 1% of people said they had one.

Q54 Lord Hope of Craighead: Good morning. My question is about the effect of voter ID on voter participation, which is dealt with in chapter 4 of your report. The figures are quite interesting. You start off with an overall figure of 8% covering all adults of voting age, but when you break it down into particular groups the figures begin to rise: 11% in the case of those from the most deprived areas in England, 14% in the case of renters, 15% for those from ethnic-minority backgrounds and then right up to 22% in the case of those with a disability that prevented them from voting in person. Before I move to my main question, why do you select renters as a particular group? What is it about them? I think the others are self-explanatory.

Gideon Skinner: Just because it was a noticeable finding, and we were looking for where there were significant differences. As I mentioned at the beginning, there is a consistent series of groups that tend to be more affected in their attitudes towards photo ID, and renters come up in some of those questions as well so it felt like it was worth looking at them again.

Dr Glenn Gottfried: It might relate to the demographics of the people who make up renters. For example, many renters are young people.

Q55 Lord Hope of Craighead: I follow. I have two questions that look behind these figures. First, to what extent do you think photo ID was the primary driver for this discrepancy? That assumes that there are other factors that have a part to play in it. Secondly, to what extent are those groups less likely to vote anyway, irrespective of photo ID? Could you enlarge on those two points?

Gideon Skinner: To give a brief bit of context, in terms of how we interpret people’s answer to that question, “less likely to want to” is not necessarily equivalent to “I definitely didn’t vote because of it”. People may be bringing in other questions—say, feelings of dissatisfaction towards the process—that might feed into that.

As an aside, some people said the requirement might make them more likely to want to vote. That included ethnic minorities, who were one of the groups who were slightly above average in saying that it made them more likely to want to vote.

On your specific questions, we can look at that 8% in more detail as we have broken it down a bit more. There is evidence on both sides. On the one hand, this is a group of people who are more likely to prioritise convenience and ease of voting over security; security is still important to them but their prime driver was convenience. A subsection of the group who did not vote are more likely to mention photo ID as one of the top reasons why they did not vote, and there is some evidence that that group may be more sensitive about voter ID.

On the other hand, as you said, we also know that that group is much less likely to vote in local elections specifically. Overall, 47% said that they always vote in local elections but, among that 8%, only 32% said that were likely to vote. That group was also less likely to have voted in the recent local elections. Although one of their more important reasons for this was voter ID, as I mentioned, they gave a range of other reasons such as being likely to be away or not being interested. They also mentioned levels of political apathy and a lack of trust in politicians, local government and so on.

So, as I say, there is a bit of both on both sides. There is some evidence that that group was perhaps a bit more sensitive but, equally, it is difficult to say that voter ID was the primary or sole reason that stopped them from voting. They were less likely to vote anyway and gave reasons besides voter ID as reasons that stopped them voting.

Lord Hope of Craighead: The answer you have just given is very much on point because the question for us is this: what weight should be attached to these figures in any recommendation that we make? Given the variety of other factors, one might think that they do not carry a great deal of weight because of the uncertainty that lies behind them. Are you able to you break those figures down for us a bit more and give us a more precise figure relating to voter ID itself? Might you be able to eliminate from the figure the people who would not have voted anyway so that we have a more accurate figure on the effect of voter ID on people who do want to vote?

Gideon Skinner: We will not be able to do it now; we can potentially go back and have a look to see whether there are any more detailed breakdowns. We will get to a relatively small sample size so there will be limits. Even so, this wider point will apply to much of the data we have: in this type of research, it is very difficult to isolate the causal impact of voter ID. We find that the groups that are more likely to mention voter ID as one of the reasons are generally less likely to vote and do still mention other issues besides voter ID. As I said, clearly, there are signs in this research that some groups might be more sensitive to the impact of voter ID, but it is difficult to narrow that down and say, “It was a purely causal thing and they would have voted had it not been for that”.

Lord Hope of Craighead: You mentioned those from ethnic-minority backgrounds where, in fact, there is a positive side to this. Could you say that in the case of people with disabilities or is there really a negative effect in that case without a positive?

Gideon Skinner: We can have a quick look. From memory, for all groups, the majority response was that it makes no difference. The smallest response tended to be people saying that it made them more likely to vote rather than less likely to vote; the net balance figure was negative but both figures were relatively low. Something like 8% said that it made them less likely to vote while something like 4% said that it made them more likely to vote. Sorry, I do not have the exact figures, but it was in that sort of range.

Lord Hope of Craighead: Would it help if we left it on this basis: if you can improve on these figures and make them a little more precise for us, you will do so in a written answer later on rather than trying to produce an answer off the cuff?

Gideon Skinner: Yes; we would be happy to do so. From a quick look, it is not that people with disabilities are significantly more likely to say that they are more likely to vote. Among people with any disability, 3% said that it made them more likely to want to vote; that is basically similar to the overall figure, which, as I said, was 4%.

The Chair: Will you take away Lord Hope’s question and refine the figures, in particular to help us see the extent of the effect of voter ID?

Gideon Skinner: Yes; we can have a look at that.

The Chair: That would be really helpful. You have triggered a pile of interest here. Lord Anderson, Baroness Suttie and Lord Falconer all have questions.

Q56 Lord Anderson of Ipswich: I have two points. The first is supplementary to what Lord Hope had to say. It strikes me that the net figure would be particularly useful. You said that 8% said “less” while 4% said “more”. It sounds as though you can produce that without getting into uncomfortably small sample sizes.

My other point relates to the most striking figure here: 22% of those with a disability that prevents them voting in person said that voter ID made them less likely to want to vote. My understanding is that you need voter ID only when you vote in person. At the moment, I do not understand. I know that it is not your job to get inside the heads of the respondents but I am puzzled. I wonder whether they fully understood this question since, if they were unable to vote anyway, it is difficult to see how the requirement for voter ID made any difference. They would be looking at proxy voting and postal voting, presumably, rather than attending a polling station.

Gideon Skinner: That is a fair point. To give the precise definition of that group, it is people who told us that their disability prevents them voting in person on their own. So it is possible that some of that group would still be able to vote but only with assistance.

Lord Anderson of Ipswich: It could simply be someone who needed to be driven to the polling station.

Gideon Skinner: Yes, but I take your point: it is slightly odd that some people in that group may well not be able to vote in a polling station in any case. That probably goes back to my original answer, in that some of this may just reflect broader levels of dissatisfaction or concern that are not necessarily related to that exactly.

The Chair: So the dissatisfaction comes out in other ways.

Lord Anderson of Ipswich: You are accepting, I think, that we need to read quite an important qualification into paragraph 4.12 of your report. It refers to “22% of those with a disability that prevents them from voting in person”; perhaps it should say, “prevents them voting in person without assistance or on their own”.

Gideon Skinner: You are right, yes.

Lord Hope of Craighead: You could use the word “impacts” there.

The Chair: Let us stay with that question from Lord Anderson. To what extent would the particular ID requirement have contributed to that figure? For example, such people may be less likely to have a passport or a driving licence. Did you pick that up in your figures? It may be about the nature of the ID. Their lack of mobility may of itself lead to them not having the particular type of ID needed. Is there anything in your figures on that?

Gideon Skinner: We will have that. I do not have it off the top of my head but we asked what types of ID people had.

Glenn Gottfried: We asked what they had and what they used, I think.

The Chair: One of the things we will look at is whether the list of IDs that would be acceptable should be lengthened, or even shortened.

Gideon Skinner: Overall, people in that very specific group—those who are unable to vote on their own in person—are less likely to have any form of photo ID. The figure for that group was 10%. They were less likely to have a passport; only just over half of them said that they have a passport, while only just over half of them said that they have a driving licence. That compares against around eight in 10 among the public as a whole. So, although most of them had photo ID, it was less prevalent than among the public as a whole.

The Chair: So the type of ID being required could contribute to that figure.

Baroness Suttie: Lord Anderson has covered the point that I was going to make. Thank you.

Q57 Lord Falconer of Thoroton: Thank you very much for coming; your evidence is incredibly useful. You said that the figures show that 8% of people said that voter ID made them less likely to want to vote in the local elections. Can you break that down by reference to age? Part of your report says that, if you are of voting age, the younger you are, the more you will focus on convenience.

Gideon Skinner: Yes, you are definitely correct about younger people’s preference for convenience over security, whereas it is the other way around for older people, for whom security is more the issue. The “less likely to vote” figure was marginally higher among 18 to 34 year-olds—12% as opposed to 8%—so there was an impact: it was slightly higher among young people.

Lord Falconer of Thoroton: What about people over 34?

Gideon Skinner: I do not have that figure. It will be similar to the overall figure. It was slightly higher among that 18 to 34 group.

Lord Falconer of Thoroton: Could you let us have whatever age breakdown you have?

Gideon Skinner: Yes.

The Chair: We are taking note of the extra questions we are asking you to take away, and we will follow up with a letter confirming, so do not worry if you do not remember everything. However, we will chase you to provide it.

Q58 Lord Hope of Craighead: I have just been looking at paragraph 4.17, which is in a different chapter, but it shows that your polling system is very sensitive to things like people who lack an interest in politics, people who rarely or never vote in local elections, and people with a disability that limits them and makes it difficult for them to get there. In the earlier paragraphs—4.11 to 4.13—why did you not break the groupings down a little more to try to identify precisely where the problems lie? You have possibly done that but not reported it in the report.

Gideon Skinner: For other groups, we pointed out that, for example, it was those who were dissatisfied with the voting process, and particularly those without identification, who were most likely to feel that they were put off voting in the locals. But I would not be surprised if it were the same—we did not want to repeat ourselves every time, or the report would end up being a long list of sub-groups. But there is definitely a consistent pattern that those who are less likely to vote in locals or generals, or who are less interested in politics, tend to feel more negative about some of these points, as you would expect.

Lord Hope of Craighead: Will you bear that in mind when you produce your supplementary information for us?

Gideon Skinner: We would be happy to add more to those groups.

Q59 Baroness Andrews: This is a related question about awareness of voter ID and the voter authority certificate, and how it varied between the different demographics. In your research, you identified aspects of those demographics, so how comprehensive was your scoping of your breakdown of demographics? For example, you pick out that awareness of the policy change was highest among older age groups—89% of those over 65—and lowest among the youngest. Would you not expect that? Was that broken down with any more granularity so that we could see, for example, some rough class distinctions between those groups? Did any aspect of gender come into your analysis? Were women or ethnic-minority women, for example, less likely to be aware?

What do you mean by “those with a greater interest in politics”? Are you simply measuring people who turn out to vote, or are you using some other more subtle indicator? Again, are we looking at educational attainment here or people who, by definition or by professional choice, would have a greater interest in politics? That question attaches to adults with photo identification too, because we are more likely to find those in professional groups—among medics and so on. On institutionalised groups, were the people who were not likely to be aware those who are not connected to a more formal workplace or are not in work at all?

This is a slightly complicated question, but I am trying to test out the extent to which what you found in relation to awareness was something you might have been able to predict, or whether there are more subtleties within these broader figures that would answer the questions raised today. How big an issue is this? Are people being accidentally deprived and disadvantaged because of the nature of their situation? Therefore, can something be done to raise that awareness, or are we looking at some very deep structural institutional barriers here in people’s attitude to politics and so on?

Gideon Skinner: I will not put my hand on my heart and say that we listed every single demographic—there is a range of demographics in the data we provided the Government in our computer tables, which include a number of those. I will not claim that we listed every single one where we saw a difference in the report.

The point about the sub-groups is that we primarily focused on the groups as a whole and did not look at interlocking them by looking at, for example, ethnic minorities or women. It may be possible to do that in some cases but, for others, we would be breaking small sample sizes down even further and it would become harder to give you results that you would have confidence in because they would have a robust number behind them.

On your point about what “interest” means and the impacts on some of the other groups you mentioned, there are clearly some signs—we have tried to pull them out on occasion—of groups such as people living in more deprived areas tending to have lower levels of awareness. We tended to find that people who are long-term unemployed had lower levels of awareness, so it looks like socioeconomic factors are playing a part.

On the specific question about how we measured a lack of interest, we had a number of different questions. We asked a simple “How interested are you in politics?” question, but we also asked a couple of questions about people’s voting behaviour in general elections and local elections: “Do you always vote in them?”, “Do you only sometimes vote in them?” and “Do you hardly ever vote in them?”, for example.

We have used a mix of those types of indicators to look at the differences by interest and political engagement. However, you are right to say that there is certainly a link between low levels of political engagement and low levels of political interest—which tends to be where awareness is lower. That feeds through throughout the research. There is something about that group. When you look at, for example, the reasons that people give for not voting in elections, it is always worth remembering that the top reasons are not photo ID but are around convenience—“I was not in” or “I was away”—or around voter apathy or lack of political engagement, distrust of politicians, not thinking it would make a difference, those types of issues. Certainly they come through as important reasons for stopping people from voting.

Baroness Andrews: Can I pursue the social and economic indicators? We had evidence from Graham Farrant, the very experienced local government official in Bournemouth. He said that they had found that there were certainly boroughs and small areas within a constituency where there was far less awareness than in other places. Did you pick up any of the geographical disparities in what you did as well? My second question, while I am on a roll, is: to what extent did your analysis match up with the Electoral Commission’s analysis? Were you consciously comparing notes as you went through? Did you find any disparities or are you looking at the same phenomena?

Gideon Skinner: On the geography one, yes, definitely. As I mentioned, one of the breaks that we have is levels of deprivation using the index of multiple deprivation figures and breaking this up into quintiles, the most deprived areas versus the least deprived areas. We find some evidence that, for example, lack of awareness is higher in the most deprived areas. That did come through.

Dr Glenn Gottfried: Also, where political interest is lowest as well.

Baroness Andrews: You picked that up in the previous question: 11% of those from the most deprived areas in England are less likely, et cetera. Within that particular deprivation index and geographical location, did you also pick up the same sort of things around awareness as you picked up from the responses to “Are you more likely to have found it difficult to vote?”

Gideon Skinner: Specifically on that, yes, awareness was lower in the most deprived areas. On awareness of voter ID, the requirement as a whole, 23% of people in the most deprived areas said that they did not know very much or anything about it. That compares with 17% overall, so there were some signs that awareness was a bit lower in those areas.

On the point about checking with the Electoral Commission, we were not talking to it specifically about its research. There may be other conversations going on between it and government, but we were not talking specifically to it about our research. Clearly, from some of the results and, apart from the point about the awareness question that you raised at the beginning, the broad pattern looks pretty similar between our data and that of the Electoral Commission. You find similar proportions of non-voters saying, for example, that photo ID was one of the reasons why they did not vote. I took the point about the awareness one at the beginning but otherwise it did not feel that there were huge differences.

Q60 Baroness Andrews: Finally, and this question may not be totally answerable, what can be done to increase awareness among these groups who are clearly not as aware as others? Do you have conversations with, for example, local government officials, who are really keen, from the evidence that they have given us, to ensure that everybody has an equal chance of being aware? Do you have any policies to promote that?

Gideon Skinner: Less so particularly from this research but more broadly, yes. I am sure this will be taken into account by the Electoral Commission and others who are carrying out their awareness campaigns. Obviously you want to carry out research and gain insight from some of those groups because you will want to target the groups where there are the lowest levels of awareness.

We asked people, if they had heard about it, what their sources were. I take it that that is not asking the people who had not heard about it, but nevertheless it can give you pointers. With younger people, one of the groups where awareness was lower, they were more likely to rely on word of mouth and social media. With the long-term unemployed, whom I mentioned were one of those groups with lower levels of awareness, it was still relatively low down the list but they were more likely than average to mention hearing about it through charities, for example, so there may be ways in which we can reach them by going through partners with other organisations. TV and radio still have the broadest reach but if you are thinking about targeting, you can think about different channels that work better for those groups that you are most interested in boosting.

Clearly, it would be difficult. You have to remember that, as was raised at the beginning, many of these groups are those with the least interest in and engagement with politics, so it would be difficult to overcome that, but there are behavioural frameworks and learnings from behaviour change in terms of thinking about the different triggers that you can use to engage people, whether that is increasing their levels of motivation, simple awareness or ensuring that the process is very simple to encourage them. More detailed research will uncover which of those aspects you really want to stress for those groups that you most want to target.

Dr Glenn Gottfried: Just to add to that, we did ask them if they had seen any of the Electoral Commission’s advertising campaign. Young people and people in more deprived areas were more likely to say that they had seen something. However, given that they are less likely to say that they know much about it, it might be shown that they are seeing it but not particularly absorbing what they are seeing. It seems that the message is getting to them but they are not taking it in, perhaps.

Gideon Skinner: To some extent, but we would have to say that the differences are not huge.

Dr Glenn Gottfried: Yes, but they show that they are not lower than overall. This is from what they are telling us.

The Chair: On one aspect of Baroness Andrews’s questions, there is significant difference between your figures and those of the Electoral Commission on particular issues surveyed. We know that there was a difference on awareness of voter authority certificates. Were there any others? Do you have a top two or three differences between your findings and those of the commission?

Gideon Skinner: I cannot give an answer to that. I have not gone through and compared the detail.

The Chair: Can we leave that one with you as well, to tell us whether there are any significant areas?

Q61 Lord Thomas of Gresford: We are concerned with voter suppression. Did you conclude that the voting had been supressed in any way and in any particularly demographic group?

Gideon Skinner: Could you explain exactly what you mean by voter suppression?

Lord Thomas of Gresford: People not voting because of the requirement of ID.

Gideon Skinner: We asked everybody who said that they did not vote in the local elections what their reasons were. As I mentioned, the main reasons were things like being away and lack of engagement. Four per cent of non-voters told us that it was to do with voter ID, which I do not think is that dissimilar to the figures the Electoral Commission got in its research.

Again, it is a smallish number, but there was broadly that split between around 2% of people who said it was because they did not have voter ID and the rest saying either that they disagreed with the policy or that they meant to bring voter ID and forgot, or for a number of other minor reasons. That was the figure that we had.

Lord Thomas of Gresford: Were there differences between demographic groups?

Gideon Skinner: Yes. Most obviously, it was highest among people without a valid form of photo ID—36%, I think. I have it in front of me. Overall, 36% of the people without photo ID who did not vote said that it was something to do with photo ID.

Lord Thomas of Gresford: That is not necessarily a demographic group, is it?

Gideon Skinner: We also found that it was slightly higher among council renters and people from more deprived areas. It was around 8% among council renters and 7% among those from more deprived areas. It was higher among those whose disability had an impact on them being able to vote in person. It was also related to some of the wider issues that were mentioned earlier; for example, it was slightly higher among those who are dissatisfied with the voting process. There were some attitudinal aspects as well as the demographic ones that you mentioned.

Q62 Lord Strathclyde: I am interested in a slightly different aspect of this, which is an enhanced list of accepted ID. There has been a good deal of debate and discussion about this. We have heard some evidence about potential changes; for instance, a gun licence is excluded, although I cannot imagine anybody who has a gun licence not having other acceptable forms of ID. In your view, and based on your survey findings, is there an argument for accepting digital forms of ID, in particular? Would that help your figures on participation?

Gideon Skinner: Unfortunately, in this data and research, we did not ask about a wider range of IDs.

Lord Strathclyde: But I wonder whether it came up.

Gideon Skinner: Not in this type of survey. We have to be quite structured about the questions we ask rather than using more qualitative forms, which you might hear about in other evidence and where there is more of an opportunity for people to come up with some of these things.

The one bit of our research that might be relevant to this discussion is that it seems clear that it is not just about availability or access. People are clearly making decisions on what ID they bring. For example, lots of people have a passport but it was not the top method used to vote; driving licences were much more likely to be used even though passports are the most widely held form of ID. This suggests that, when people are making decisions about this, it is not simply about access to or availability of ID; they may be thinking about other issues. I am speculating but it might just be a matter of convenience—that is, whether you have the ID on you—or there may be concerns around security and other aspects. In terms of IDs that people actually use, it is not simply about access; they are making judgments about other issues as well.

Lord Strathclyde: I do not have anything to come back to you on that. Basically, you are saying that you did not ask the question.

Gideon Skinner: We did not, I am afraid.

Lord Strathclyde: It is therefore difficult for you to give an opinion on that.

Q63 Lord Foulkes of Cumnock: I have a follow-up question. ID is being asked for on more and more occasions. I was surprised to be asked for ID the other day in flying to Edinburgh. Thankfully, the Scottish nationalists have not imposed that on us yet, but the airline seemed to do it. Are you aware of whether more people and organisations are asking for ID for travel or other purposes?

Gideon Skinner: Not off the top of my head, I have to say. I am not aware of research around that. We can perhaps have a dig. Clearly, the use of different forms of ID is changing. There are different uses for digital ID and so on but, again, I do not have anything on the implications of that being used in this form or for this purpose.

Lord Foulkes of Cumnock: It would be interesting to see whether the list of documents or other forms of ID that can be accepted by organisations introducing ID as a requirement for travel or any other purpose is different from the list for voting; that includes whether it includes the kind of suggestions that Lord Strathclyde made about digital and other IDs. You do not know as you have not done anything on that, yes?

Gideon Skinner: Certainly not in this survey. I am not aware of anything else on that; nothing immediately comes to mind. I refer back to one of my earlier answers. If you are thinking about what behavioural insights might tell us in terms of encouraging people to become aware of the requirement and the types of ID they may use, the wider social context is clearly something that you would want to bear in mind. What their friends, family and neighbours are doing is likely to have an impact on people.

The Chair: Finishing off Lord Strathclyde’s point, obviously, you asked questions of people and produced evidence on the extent to which the requirement for voter ID influenced people’s disposition or presented a barrier to their participating in a vote. Did your questioning allow people at any point to distinguish between the barrier being posed by the particular ID requirement list as opposed to ID generally? Is there anything there that would help us think about the pros and cons of extending or adding to the list?

Gideon Skinner: It did not, I am afraid. When we asked questions, it was about the requirement in general. We asked basic behavioural questions about which IDs people had and which ones they used if they voted; earlier, I made the point about people using driving licences more. However, we did not get into the granularity of types of voter ID beyond that.

The Chair: Of those who did vote using ID, you asked what form they used. That is fine—basically, you do not have the evidence to give us that kind of information.

Gideon Skinner: I am afraid not.

The Chair: That is fine.

Q64 Lord Keen of Elie: Good morning, gentlemen. You may already have dealt with this point but some additions to voter identification have been suggested, such as attestation by another registered voter or a statutory declaration by the voter themselves. Does anything in your research indicate whether either of these methods could prove voter intentions?

Gideon Skinner: Even less so than the previous question, I am afraid. We just did not ask about attestation, vouching or statutory declarations in this survey. I am sorry.

Lord Keen of Elie: I rather anticipated that answer but thank you anyway.

Gideon Skinner: It is worth pointing out that this survey was not just about covering attitudes towards voter ID; it also covered other aspects of the programme, such as the changes to support for people with disabilities. It was a long-ish survey but it covered a number of topics.

Lord Keen of Elie: I quite understand.

Q65 Lord Thomas of Gresford: I appreciate that it is not your job to be a seer as to the future and that your expertise is, in essence, gauging what has happened in the past or in the present. However, what impact do you anticipate the voter ID requirement having on turnout at the next general election? Do you think that it will vary between different demographics?

Gideon Skinner: Speaking as someone who also sometimes does a lot of political polling, I wish that other people held the same attitude as you towards us being seers. I know that my answer will probably be similar to that given previously by some of your other witnesses. It is very difficult to give a forecast for turnout. Our data is on local elections. We know that the groups of people who vote in local elections are different from the groups who vote in general elections, so the impact might be different either way. We also know that there are so many other factors that go on in turnout.

Going back to the earlier point about what prevents people, it is general overall attitudes to the Government, to voting and even to the time when it is happening, as well as the structural points raised earlier around a lack of engagement or interest in politics. We would expect people to be more motivated to vote in general elections, so it is slightly different. It is very difficult to give an answer about what might happen, particularly a year or so in the future.

From this dataset and the wider research, we know that, in the local elections, the change had a relatively small but still measurable impact, and some groups seem to be more sensitive towards this. But, at the same time, it is difficult to say, from this data, that voter ID was the only cause of that, because those groups tend to be less likely to be interested in politics, less engaged and less likely to vote anyway. That is as much as we can say.

Lord Thomas of Gresford: I used the phrase “voter suppression” in my previous question. Do you fear voter suppression as a result of the requirement for voter ID?

Gideon Skinner: I do not have an answer to that, based on the data we have. As I said, we know from the data that some groups are more concerned about it or think it might make it more difficult for them, but most people say it would make no difference. The former is most obviously higher among people without photo ID. But we know that there are other options, such as postal voting and so on—some of these groups are more likely to vote by post already—so there are those other aspects. We are talking about voting in polling stations, but there are other options as well.

Lord Thomas of Gresford: Is it your opinion that the result of the next election is unlikely to be affected by the requirement for voter ID?

Gideon Skinner: It is very difficult to say. This would be purely speculative and not based on any evidence. Given the numbers, I would broadly agree that it would be unlikely, but it is difficult to say whether it would have a significant impact.

Q66 Lord Falconer of Thoroton: You say in your report that 21% say that the requirement makes it more difficult to vote in person, but 10% say it makes it easier. I can see that some people might have more confidence in the system with the ID, but how does it make it easier to have to give photographic ID?

Gideon Skinner: I take your point, but I suppose it may be related to your thinking about how people may have answered this question: for example, people feeling that ID makes them more confident may encourage them in some way, and that may partly drive their answer. They may think it is simpler and that they just have to show someone something that is official, and that might be informing their answer.

Lord Falconer of Thoroton: Tell me if you cannot reach a conclusion on this, but is your overall conclusion—without putting a scale on it—that the requirement for photo ID will probably mean that fewer people vote in a general election, even if you cannot say what the number is?

Gideon Skinner: I am finding it difficult to answer because I feel I have little evidence of that. There are some people who think it made a difference for them in the local elections, but, as I said, we know that some of them are less likely to vote anyway and have other reasons for not voting that are not simply around voter ID but involve a lack of engagement in politics and so on.

Lord Falconer of Thoroton: Your polling suggests that 8% of adults of voting age said that it made them less likely to want to vote in the local elections, although 4% said that it made them more likely—so that is more people not wanting to vote. You have that figure of 21% versus 10% in relation to saying it will make it more difficult. You also have figures saying that, if people forget their ID, although perhaps 60% would say they would go home to get it and come back, 30% or 40% said they would not. Do those figures not suggest that, ultimately—without putting a scale on it—fewer people will vote in a general election?

Gideon Skinner: I take your point. From my understanding of the data that was collected by the Electoral Commission in polling stations, measuring how many people came back, about one-third of those who were turned away did not come back—very roughly speaking. These are all smallish numbers, so I do not think that number was completely different. So, yes, that is consistent with that.

You mentioned people being less likely to vote in local elections and thinking about how that might apply in general elections. You could see that people who vote in general elections are a different group and will be less regular voters in elections generally and probably a bit less interested in politics generally, so this may have an impact from that point of view. But, equally, they are voting in the general election for the very reason that they are more motivated to do so, which might play in the other direction. Clearly, there would need to be awareness-raising campaigns and so on around it ahead of the general election.

But, as you can partly tell from the Electoral Commission research, there seems to be a small group of people who were turned away and did not come back—so they were affected. There is also a relatively small group of people, particularly those without photo ID, who say that this makes them less likely to want to vote and is one of the contributors to their decisions not to vote. So, all else being equal, the best educated guess you could make would be that that would not be terribly different in a general election—but that is educated guesswork as much as having any data.

Lord Falconer of Thoroton: On the basis of your research and information, are you willing to make any estimate of the figure for the reduction in the number of people who might be willing to vote in a general election as a result of the voter ID requirement?

Gideon Skinner: No, I cannot give a number, partly because we did not ask questions on that. With a year before the next general election, we know that things can change and that, even if we are collecting the data in the most perfect way possible, people’s judgment and knowledge of what will affect their behaviour, especially a year ahead, will be fallible.

The Chair: Staying with Lord Falconer’s questions, you referred to the 8:4 split between it being more difficult and less difficult, and the 21% figure regarding less likely and more likely. At the aggregate level that is a small group of people, but would it be correct to say that it is not so small in relation to the population of a particular type of demographic? If you were looking at disabled people then you could not necessarily say it was a small group based on your figures combo or if you are looking at the unemployed then it is not necessarily a small group of the unemployed.

Gideon Skinner: That is correct. It is worth pointing out, in terms of that “difficult” figure, that we asked it in two ways: we asked overall, generally speaking, “Do you think it would make it easier or more difficult for you to be able to vote, or would it make no difference?” That is where the 21% figure is coming from. We also specifically asked people in the local election areas whether it made it easier or more difficult for them to vote in the local elections, and we got a slightly different figure there. When we asked specifically about their experience in the local elections for people in local election areas in England, 12% of people said it made it more difficult for them to vote, 5% said it made it easier and most people said it made no difference. There is still clearly a negative balance—more people said it made it difficult for them—but not quite as large as when we just asked people to think generally.

Yes, that figure is higher among certain groups, most notably among people who do not have ID, as you would expect. It is worth pointing out that this question was about making it more difficult to vote in person at a polling station, rather than voting generally. It was higher among those people who do not have ID, nearly half of whom said it made it more difficult for them to vote; among those with a disability that impacts on people’s ability to vote in person; among younger people; and among ethnic minorities, although ethnic minorities were also a bit more likely to say it made it a bit easier for them. On deprivation, it was higher among those in the most deprived areas, although they too were slightly more likely to say it made it easier to vote. So it follows some of those groups.

The Chair: So it is not necessarily so small when you look at particular demographics.

Lord Anderson of Ipswich: Those figures are at paragraph 4.10 of your report. I have another tedious question about your questions to people who are disabled. You have just used another formulation about people who are disabled, which indeed appears at paragraph 4.10: those with a disability that impacts on their voting in person. You told us that, despite what it says in paragraph 4.12, you asked that question about people with a disability that prevents them from voting in person on their own. Those could be subtly different. If you chose a different form of words for those two questions, could you explain why?

Gideon Skinner: There are a couple of different questions that we asked people with disabilities—again, remembering that part of the focus of this research was to cover the changes to support for people with disabilities in voting, having companions and more support available in polling stations. We wanted a couple of different ways in which we could speak to them and understand their views.

We asked a general question about how much their disability limits their activities, which was on a simple scale between “A lot”, “A little” and “Not very much at all”. Then we asked a question specifically about the impact on voting in person, which had a scale from “It prevents me from voting in person on my own completely” to “It has an impact but it does not prevent me” and then “It has no impact at all”. The figure that I just quoted there was among the group who feel it has some impact, so they are likely to have a more severe disability but not the most severe.

Lord Anderson of Ipswich: So when it says 18% of those whose disability impacts on their voting in person, that does not include those who are prevented from voting in person on their own—or are they included in the category?

Gideon Skinner: No. They are the middle group.

Lord Anderson of Ipswich: So it is only the middle group here. If you added in those who could only vote in person with someone else, would that figure be different?

Gideon Skinner: Your questions have not been tedious at all, but I worry that my answers might have been. This is partly a technical point about the sizes of these different groups. For example, the other group that you mention, the group who say it prevents them completely from voting in person on their own, also had a higher proportion of people who said they found it more difficult, 17%. That is very similar to the other group but, because they were a smaller base size, strictly speaking that was not statistically significant. That is why we did not mention it in the report.

The Chair: There is one curiosity. In your findings regarding the introduction of photo ID making it more difficult, we have graduates at 17%. Where did that come from?

Gideon Skinner: I agree with you that it seems a bit odd, because elsewhere we know that they are more interested in politics and so on. I do not have a precise answer for that, but it was a result that came through so we felt we should mention it. It may be that it is partly related to age because graduates tend to be a bit younger and we know, as we see in the same para, that the under-54s were more likely to find it more difficult, so it may be that there is partly an age aspect there and that is more the driver.

Q67 The Chair: If no one else has any more questions, I have one overarching question. Obviously, we are interested in all parts of the UK, and you surveyed Northern Ireland. Were there any distinctive features of the evidence regarding the party of the survey conducted in Northern Ireland?

Gideon Skinner: We did not ask the questions about voter ID in Northern Ireland. In some of the other questions, we found some differences in Northern Ireland—for example, the attitudes to postal voting were different. There was more negativity to postal voting there than we found elsewhere. But otherwise, general attitudes towards voting and general levels of confidence in the voting process were similar. Among those who were unhappy in Northern Ireland, issues around coercion came up a bit higher. They were not hugely different, but there were some differences around those two aspects. However, we did not ask about voter ID there; we just asked about voter ID in England.

Lord Foulkes of Cumnock: England? Not Scotland?

Dr Glenn Gottfried: Not in wave 2. We asked in Scotland in wave 1.

Gideon Skinner: Yes, in our first wave we asked in Scotland but not in the second wave, because the locals were happening in England.

Lord Foulkes of Cumnock: Ah. Sorry.

The Chair: I think that is all our questions. Thank you. That was very helpful.

Q68 Lord Falconer of Thoroton: I am sorry to interrupt, but I have one more question. What were you tasked by the Government to do?

Gideon Skinner: They issued a brief saying that they wanted to commission an independent survey to feed into the evaluation of the electoral integrity programme. I cannot remember exactly how it was issued, but it would have been open to others to bid for it, and it would have had its objectives: “These are broadly the areas that we want to cover. These are the groups that we might be particularly interested in, so the sample size needs to be big enough to be able to look at people with disabilities, for example, or in different areas, so that we can look at that”.

Very basically, we put in a proposal for that, as did others, and they chose us. We would have had our standard set-up meeting and discussion with them. We would have had drafts of the questionnaire going back and forth. Then we would have gone into field, and then we would have reported.

Lord Falconer of Thoroton: Sure. There is nothing sinister about the question. For reasons to do with I do not know what, I do not have the first few pages of your report. This is my fault, not yours. Under the heading “Introduction and methodology”, do pages 5 to 9 set out what your instructions were and the questions you answered? That might answer my question to you.

Gideon Skinner: Very basically. It gives a brief introduction and the background to the programme, and then more detail about the methodology we used, some of the key aspects covered by the questionnaire, and the aim of the policy measures. So, yes, it gives a brief bit of background, and then it gives more detail about the methodology.

Lord Keen of Elie: I think Lord Falconer may discover that the issue is with pagination, not content.

Lord Falconer of Thoroton: Ah. I have page 4 and then page 9.

The Chair: You have an edited copy, I think. I looked at it last night. It was circulated under an earlier edition.

Lord Falconer of Thoroton: It is my fault.

The Chair: No. It sets out the four aspirations of government, I think.

Lord Anderson of Ipswich: I am in the same position as Lord Falconer; I have an edited version of the report. Calling it up on screen, the only sentence I can find that is relevant to what they actually asked you to do is in paragraph 1.1, which says: “The aim of this research is to understand the public views towards and experiences of voting, and the potential impact of the Government’s electoral integrity programme contributing insight towards the evaluation of this important set of policy measures”.

I do not know whether you received any more detailed brief from government that is perhaps not directly quoted in the report.

Gideon Skinner: There was a more detailed brief. Without passing the buck, I suspect that you can ask DLUHC for that.

The Chair: Maybe we can come back to you on that. I think, from memory, that there was more detail in the introduction about the areas the Government were interested in, or the aspirations they had—what they were seeking to achieve. We can look at that and come back with any questions. We will get that clarity for you, Lord Falconer.

Lord Falconer of Thoroton: Thank you.

The Chair: Thank you very much. That was very helpful. We will have to go through your evidence again in the light of what you said to us.