Ep 312: UKGC wisdom in short supply, statistically speaking

The UK Gambling Commission has published its first-year statistics and datasets of the Gambling Survey for Great Britain for 2023. 

Replacing the previous datasets provided by the NHS on problem gambling prevalence, the survey is a key directive of the commission, which in 2021, launched a pilot programme to re-model how data is collected on adult gambling participation and problem gambling prevalence statistics. 

And this will be the topic of today’s episode of iGaming Daily, supported by Optimove, where James Ross is joined by Melanie Ellis, Partner at Northridge Law, to talk about the new survey and key highlights from the data provided.

To read more on the topics discussed in today's episode, click on the following links:

- https://sbcnews.co.uk/europe/uk/2024/07/25/ukgc-new-baseline/
- https://www.gamblingcommission.gov.uk/report/gambling-survey-for-great-britain-annual-report-2023-official-statistics/gsgb-annual-report-additional-adverse-consequences-from-gambling
- https://www.gamblingcommission.gov.uk/news/article/first-gambling-survey-for-great-britain-annual-report-published
- https://bettingandgamingcouncil.com/news/bgc-statement-on-gsgb-publication

Host: James Ross
Guest: Melanie Ellis
Producer: Anaya McDonald
Editor: James Ross

Remember to check out Optimove at https://hubs.la/Q02gLC5L0 or go to Optimove.com/sbc to get your first month free when buying the industry's leading customer-loyalty service.

Creators & Guests

Host

James Ross

Multimedia Editor at SBC Media

What is iGaming Daily?

A daily podcast delving into the biggest stories of the day throughout the sports betting and igaming sector.

James Ross (00:17.019)
The UK Gambling Commission has published its first year statistics and datasets of the Gambling Survey for Great Britain for 2023. Replacing the previous datasets provided by the NHS on problem gambling prevalence, the survey is a key directive of the commission, which in 2021 launched a pilot program to remodel how data is collected on adult gambling participation and problem gambling prevalence statistics.

And this will be the topic of today's episode of iGaming Daily, supported by OptiMove, where I'm joined by Melanie Ellis, the partner at Northridge Law, to talk about the new survey and key highlights from the data provided. Hi Melanie, you okay?

Melanie Ellis (00:56.65)
Hi James, thanks very much for having me on the podcast.

James Ross (01:00.203)
No, no, it's an absolute pleasure to have you on. And I think it's the perfect topic for you to have you on the podcast as well. No pressure, no pressure. But before we jump into the conversation, it's time to mention the sponsor of the iGaming Daily podcast, who have an important statistic of their own because they are the number one CRM focused marketing solution for the iGaming market. And as a special offer, Optimova offering new clients a first free month.

Melanie Ellis (01:07.507)
Yeah, actually.

James Ross (01:29.593)
when they buy OptiMove for information and to claim them for free month, go to optimove .com forward slash SBC. Links will be left in the podcast description below. So back to the gambling survey for Great Britain. And I think it's important to highlight that this survey, it collects information from a sample of the population, which is designed to represent the whole population of adults aged 18 years and older living in private households in Great Britain

accurately as possible with practical constraints such as time and cost. And this according to the UKGC is why the statistics are estimates, and I emphasise that point, estimates rather than precise figures. So they could be subject to a margin of error. So Melanie, I'm going to kickstart this conversation off the back of your LinkedIn post. With these being estimates, do you think it's wise from the UKGC to release

potentially unreliable data as official statistics.

Melanie Ellis (02:34.284)
I mean, think, I mean, in short, now, I don't think it's wise for them to release these official statistics at this time. But the issue isn't just about the fact that these can only ever be estimates of problem gambling rates and the other data points that they've collected. There are concerns that have been expressed about the survey methodology after the previous

pilot stage, the experimental stage of the survey, a review was carried out by Patrick Sturgis, who recommended various steps were taken by the Gambling Commission to sort of check the accuracy of the results that were being obtained, particularly in view of what he mentioned as the potential for bias in terms of the people who would be.

taking the survey and the impact that that could have on the results that were obtained. So there are still steps the Commission need to be taking to establish the reliability of the data. So really my view is until they've completed those steps, they shouldn't be releasing these as official statistics. It's serious topics that we're talking about here, particularly problem gambling rates and even suicide is discussed in this.

data so I do think the Commission needs to treat this seriously as it deserves and take the necessary steps before, as I say, before releasing these as official statistics.

James Ross (04:12.691)
And what were your kind of key takeaways from the statistics? I know it's a big document, it's in depth, but mean, just from an overview, what were your kind of key takeaways from it?

Melanie Ellis (04:22.641)
Yeah, yeah, I'm in

Yeah so, and this is a of a difficult question firstly because of the issues that I've mentioned we don't really yet know how reliable the data is so it's difficult to actually take anything away from it. The key takeaways I think really will come in time as we have further repetitions of the survey in subsequent years and we can start to see if any of the figures are changing over time and that will give us some interesting

results. For me I say at the moment the most interesting aspects of the survey are the breakdown of problem gambling rates by the different gambling activities and I'm sure we can get on to talk about those with the caveats that the gambling commission has released in its guidance which I'm sure we'll also go through before we get into that discussion.

James Ross (05:19.267)
We definitely will. But before we do that, I think it is kind of crucial to talk about this new methodology, which the gambling commission have implemented within this new survey, because it is different. And as I say, this is the baseline study using this methodology. It's the push to web mix mode design. Can you just explain what this methodology is, how it works, and how it differs from those of the NHS in the past?

Melanie Ellis (05:46.912)
Yeah sure, so in the past the Gambling Commission has done its own surveys but these were conducted over the telephone so they telephone interviews that carried out a particular difference with those as it just used a mini screen for the PGSI with just three of the questions about gambling behaviours. I also understand this survey was becoming

expensive for the Gambling Commission to continue conducting. The NHS health survey has also given us data on problem gambling rates that used the full PGSI screening with nine questions and the methodology there was a self -completion paper booklet and it was part of an overall survey on people's health rather than just being

gambling survey. So looking at the the new survey methodology by the gambling commission, they are now conducting this by sending letters to a random selection of households and they ask two adults from the household maximum, based on whoever had the most recent birthday, to complete the survey and they're encouraged to do it online, that's why it's called push to web, but there's also the option

of sending in a paper survey and the point of that option is to try and reduce the bias that would be caused by just capturing people who are comfortable using online survey methodologies. So, so this latest iteration of the new survey version, this one that's now the official statistics had, I think, just under 10 ,000.

people participate and the Commission hopes that in the future they'll get that up to 20 ,000. I would say a key difference between this and the health survey, which was the previous source of official statistics, is the fact that this is now a gambling specific survey rather than just some questions as part of a health survey.

Melanie Ellis (08:05.696)
The risk here, which was highlighted in this report I mentioned from Patrick Sturges, is a risk of self -selection. So it's people who maybe have a particular interest in gambling, who maybe gamble on a regular basis, who are going to be more interested in going on to complete a survey which is just about gambling. It's quite a long survey as well, so it's not something people would be as likely.

to want to do unless they were engaged in this topic in the first place.

James Ross (08:40.175)
I think kind of leads nicely onto the next question I've got for you. And you've already said this was kind of a more focused. It could be seen as a more focused approach because it's focused on gambling, because it's the gambling commission. Was there a need to actually change the way that data is collected?

Melanie Ellis (09:00.052)
Yeah, so I think you can understand the reasoning for the Commission wanting to move away from a telephone survey. That's going to limit the number of people realistically they can sample and the length of the survey was limited as well. And in terms of the NHS survey, that was regarded as the gold standard, but I think from the gambling Commission's perspective, they didn't have the control.

over the survey that they wanted so they were reliant on when the NHS chose to put the gambling questions in the survey and they're reliant on them for the questions that would be included whereas now the Commission can decide what they want to ask about and when they want to ask

James Ross (09:48.673)
Mm -hmm. And I know the gambling commission stated that you can't compare these results to other surveys. And we're going to do our best not to compare them. But there are other surveys out there that we can look at similarities or differences at least. between the two, are there any differences or similarities in some of the key data that you've noticed?

Melanie Ellis (10:13.144)
Yeah, so I think we need to comply with the commission's guidance. We need to make it clear that we are not taking from this that there has been a change in rates of problem gambling or any of the other data. I don't think it's unreasonable for us to say that the NHS Health Survey found a problem gambling rate amongst the general population of around 0 .3 percent, whereas this new

gambling survey for Great Britain has found a rate of around 2 .5 % of the population. So this is where the Commission don't want us to compare these two figures. But of course, it's impossible not to notice the difference. It's almost a tenfold difference in the number of people who are found to be experiencing problem gambling behaviours.

Yeah, you know, I think all I would say about it is it highlights the fact that one or other of these surveys was not accurately finding the rates of problem gambling. And we, at this stage, I don't think we can confidently say which one it is. It may be that the true figure falls somewhere in between those numbers.

James Ross (11:40.859)
Yeah, I mean, when I was looking into this survey and it was on suicide rates, we'll talk about, suicide percentage, sorry, which we'll talk about later, the gambling commission did say that their percentage compared to the adult psychiatric morbidity survey in 2014, they say their estimates could be higher than average. You are the adult psychiatric

morbidity survey could be lower than average. So again, it goes off your point that is there's not really that much clarity or accuracy. I know that's estimates, so we shouldn't be seen as though, but it does kind of leave a lot of uncertainty. And I kind of want to jump on to the PGSI.

Is that story guided as the best measure of problem gambling prevalence on participants?

Melanie Ellis (12:35.16)
Jo, I think it is. Everybody acknowledges, I think, that the PGSI is an imperfect tool. It's certainly not intended to be a clinical tool for diagnosing gambling addiction, for example. But it's really difficult to measure problem gambling rates. And realistically, it's something that can only be assessed through self -reporting.

the indicators that are used are mostly quite open to interpretation. So it's, you know, they're going to be interpreted differently by different people and they can't give us a completely accurate reflection of what people are actually going through or experiencing. So I'd say it's an imperfect tool, but it's probably the best that we have at the moment to get some sort of estimate of rates of problem gambling.

James Ross (13:34.233)
And I do want to shift over to slots because slots were mentioned. And obviously we've mentioned that this is more gambling focused. Online slots has been earmarked in the report with a proportion of participants with a PGSI score of eight or more was more than six times higher for those who had gambling online slots. Related to all people who had gambling in the past 12 months. Is this a cause for concern?

for slot suppliers and operators in the UK

Melanie Ellis (14:09.012)
you know I think it is a concerning figure and again we have to bear in mind that the gambling commission has said that we shouldn't be yet using these figures to sort of estimate problem gambling rates in the population or you know even for slots so I think it's more something we should use over time to assess whether you know particularly to assess whether interventions like a stake limit on slots is is reducing those rates at all.

And also, I mean, when using it with caution, we have to bear in mind that correlation doesn't equal causation. So these results don't tell us that online slots have caused 25 % of users to become people experiencing problem gambling. It might be telling us that this is a product that's more likely to be used by somebody in that situation.

I think a lot more research really is needed into that. But I agree the headline figure is concerning.

James Ross (15:20.251)
For now, we'll call it a baseline concern. Maybe they're moving forward.

Melanie Ellis (15:23.168)
Yeah, think we look at it as a baseline. It may be. We may find that as the gambling commission makes some improvements to the survey methodology, we may see the figure changing.

James Ross (15:39.611)
And we already kind of, well, I already touched on the adult psychiatric morbidity survey in 2024. So we're going to kind of return to that. I already said the data for the APMS, I'm going to use the abbreviation from now on, is lower than the average given by the gambling commission. I know you have the NHS figures.

But I'm going to ask this anyway, could there be an argument that this actually from the gambling commission is bad data and an anomaly due to the sample that they've obtained?

Melanie Ellis (16:19.488)
I this is very difficult, question. The question in the gambling commission survey was about suicidal thoughts or suicide attempts and this is probably something that's very much, I'm not a statistician, I don't know for sure, but I would think this is something that's very much affected by how you ask the question and what exactly.

you're asking people when you're talking about suicidal thoughts. So I would be very reluctant to try and make any comparisons between this data and other surveys. What I found online was a rate from the NHS that 20 % of the adult population had had suicidal thoughts, which is obviously quite a bit higher than the 11 .4%.

that was found by the Commission, but yeah, as I say, it may very much depend on exactly what question was asked and maybe in what context as well.

James Ross (17:26.204)
I've only got two more questions, so we'll round it off. Like I said, this is a big, big survey and we could talk. I wish we could delve into it more within the 20 minutes that we have. But do you think that the research into pop and gambling harms, it's still underdeveloped, sorry, it's an underdeveloped field in comparison to other addictions such as drugs and alcohol?

Melanie Ellis (17:50.808)
Certainly I would agree that more research is needed, particularly I think into the causes of problem gambling and gambling harms because the figures that we have show us some correlations, which as we've said we should treat with caution, but it's not telling us what's causing people to suffer these harms in the first place or how we can prevent it, how we can help these people. So certainly...

yes I'd like to see a lot more research in this field.

James Ross (18:25.954)
And I asked at the start of the podcast what your key takeaways were from the report. So I'm going to round it off in a similar way, but what should we take away from the UK GC's figures and what can we expect in the next wave, which I believe, and correct me if wrong, it's September this year?

Melanie Ellis (18:46.816)
yeah, I'm not sure actually on the data. I'll have to take your word for it. You can, you can amend it. Yeah. so I mean, look, I, I would probably discourage anyone at this stage from taking too much away from the data. I would like to see the gambling commission complete those, those tests, additional research.

James Ross (18:53.582)
If I'm wrong, I'll make a declaration on social media. Yeah.

Melanie Ellis (19:14.742)
recommended by Professor Sturgis before really anyone is to start using this data to particularly to inform gambling policy. We also need to bear in mind, I think, that the breakdown data for different gambling activities and the rates of problem gambling is supposed to be used.

And at some point, I'm not sure when and if the gambling commission will transition from the NHS data to this new data, but it's supposed to be used by online operators to inform the number of customer interactions that they're conducting. So, I for example, for online slots, if this new data is to be used, they will need to go from

a sort of a target figure of 8 .5 % of customers that they're interacting with to 24 .5 % of customers. That will have a significant impact on a lot of operators, I would say it would mean they'll have to. I would imagine employ new people to the team that's conducting the customer interactions. So sorry, I'm not sure where we're going with this. Or should they take away?

Yeah, should people be taking away anything from that data, it will depend in that respect as to when and whether the gambling commission makes that transition from the NHS data to this new data for those customer interaction rates. So other than if it were mandated by the commission for that purpose, as I say, I would be cautious about taking it away to use for any other.

as a policy making purpose.

James Ross (21:14.179)
I caution is the key takeaway from this survey at the moment. Melio?

Melanie Ellis (21:19.274)
Yeah, for sure. But, you know, I think it's inevitable that we will see media reports just taking some of these headline figures despite the gambling commission's guidance and there's probably very little the commission can actually do to stop that from happening.

James Ross (21:42.202)
Perfect. Melanie, that's all the time we've got on the podcast today. Thank you for talking to us about this, gambling survey. And I look forward to speaking to you maybe at the, when the second wave's announced as well. So we can do a bit of a follow up from this.

Melanie Ellis (21:55.628)
Yeah, for sure. Thank you very much for having me.

James Ross (22:00.44)
Thank you. To the listeners out there, I will leave links in the description below for any relevant articles from SPC's roster of news sites, along with a link to the gambling commission's survey as well, so you can look at it yourself. Apart from that, I've been James Ross. I've been joined by Melanie Ellis, and this has been iGaming Daily.