Entering the experimental phase: Gambling participation and prevalence project
We have reached the next stage in our project to improve the way we collect data on gambling participation and prevalence of those who experience difficulties or harms through their gambling.
Posted 24 October 2022 by Communications
Following the pilot, we have now entered a 12 month experimental phase. The experimental phase allows for the continued development of the research methodology. Through a series of experiments we will be testing out different sets of conditions to further refine our methodology and questionnaire design.
This will build on the results of the pilot survey and feedback we have had from stakeholders. In this blog post, we describe what those experiments are. The quality of our statistics is really important to us. Improving the quality and timeliness of our gambling participation and prevalence of problem gambling statistics will ensure we have better evidence to help us understand consumers better and inform effective policy decisions.
Experiment 1
Household selection and presentation of harms statements. In Experiment 1 we want to explore the best ways to encourage both gamblers and non-gamblers to complete the survey to ensure a representative spread. This was a key recommendation from the pilot survey and something stakeholders rightly felt strongly about. It’s important that we get the make-up of the sample right.
One of the ways we will be doing this is to change the wording on the invitation letters to make it clearer that we are interested in the responses from both gamblers and non-gamblers, and the value that their responses can have. Feedback from our lived experience advisory panel also stressed the importance of reassurance around confidentiality of responses, so we have built in some additional wording.
We will also be conducting a split sample experiment to find the optimum approach for inviting adults in the household to take part in the survey. We’ll compare asking all adults in a household to take part in the survey against asking two adults to take part in the survey and assess the impact of these two approaches on the profile of adults who respond to the survey.
As part of Experiment 1, we’ll also be continuing to build on the work we have been doing to develop a way of understanding the incidence and nature of harms associated with gambling. We asked some questions in the pilot survey about gambling related harms, and now we’ll be running an experiment to determine whether a binary response or a scaled option works best when asking about experiences of harm. In this experiment, we’ll collect responses from a sample of 6,000 respondents across Great Britain, which will allow us to robustly evaluate which approach works best. We plan to include questions about gambling related harms in the survey on an ongoing basis in the future, so it’s vital that we get the question wording right.
Experiment 2
Construction and presentation of the gambling activity list. In the second experiment we will develop and test a refined gambling activity list as well as testing different ways to capture and present information about gambling participation within the survey. Stakeholders unanimously agreed that the gambling activities we ask about need updating. In this experiment we’ll be comparing two different ways of asking about gambling participation as well as testing a new activity list with a sample of 4,000 respondents across Great Britain.
Feedback from stakeholders also stressed the importance of collecting information about the amount of money people spend on gambling. Whilst we hadn’t initially planned to collect expenditure data via this survey, the strength of feedback amongst stakeholders about this topic, means we have decided to test an approach whilst we are in the experimental phase. This will help to inform if and how we collect this sort of information in the future.
Finally, we also plan to test the inclusion of a QR code on the invitation letter in Experiment 2. A QR code is designed to make it easier for those deciding to complete the survey online to access the survey. We’ll be interested in the impact of including a QR code on response rates and the profile of people who respond.
Outcomes
Together the findings from the two experimental steps will allow us to further refine the methodology, ensuring that we have the most robust methodology and survey content moving forward. We’ll be reporting the findings from the experiments outlined above in February next year, before we move into the third Step of the experimental stage which will be a soft launch of the survey.
We are ambitious about improving the quality of our statistics. Ultimately we are aiming for the gambling survey, alongside other evidence, to help us to build a better understanding of consumer gambling behaviours and in turn make effective policy decisions to make gambling even safer.