 Yn y cyfnod yw'r cyfnod yw'r cyfnod ar y cyfnod y Prifysgol Ffringl Nwyl, wrth gwrs, pan gwrs Catherine Grant. Catherine wedi'u gwneud o'r ysgrifennod y Prifysgol Ffringl Nwyl o Unes, neu i'r gweithio'r gweithio'r gweithio'r llwythfawr dda o'r pandemig, i'r adnod o gweithio'r gweithio'r lleceflu'r telefon o'r gweithio'r gweithio'r model hefyd. Felly, mae'n meddwl yn gweithio'r gweithio'r gweithio o'r gweithio'r gweithio. Yn yr yrwyl yw Ynrwyng Ymchwil 5 gan yw Gweithrecht Cifidus ac'r wrth gweithrecht? Fi yw Gweithrecht Cifidus? Felly, mae waith i wedi eu ffaith o'r gweithrecht Cifidus yn gweithrecht. Felly, rwy'n cwrs mai ysgrifennid fydd ar y gweithrecht llwyddiadau ar ddiweddau cyfaintfeyddus cydweith. Rwy'n meddwl â'r flynedd â'r gweld sydd yn gweithrecht, fydd â'r arddangos hwn o gweithrecht llwyddiadau'r gweithrecht gweithrecht. I want to leave some time for questions and feedback on what you find useful in the survey, any comments on the survey. So I'm going to do a quick introduction for the survey for anyone who's not familiar with the crime survey for England and Wales. Look at our return to face to face fieldwork, quick look at the CSEW result and then availability of the data because obviously that's what a lot of people are interested in. So crime survey for England and Wales is a victimisation survey with a long history, 40 plus years now of measuring crime in the country. We're really proud of that trend data, it's really important to the survey and keep maintaining that trend underpins quite a lot of the rationale for how we change the survey, how we make sure we can protect that trend, but also making sure that we're up to date and we're reflecting new crimes, we're reflecting a changing society and we're really sort of keeping up to date with user needs. We have two main measures on the survey really, incidents and prevalence, we tend to lead with incidents measures, but the survey does so much more than that and I just wanted to list out a few of the other areas that we cover here. So we're looking at, as I said, incidents and prevalence of crime, the experience of domestic abuse as well, fear of crime, worries about being a victim of crime, experiences and perceptions of antisocial behaviour, crime prevention and security measures that people take, attitudes towards the police. We've got in the questionnaire when we went back to face to face fieldwork in October 2021, we've got a whole set of questions now on satisfaction with the police. So whether or not satisfaction when you call 999, if you've been a victim of crime, if you've been a witness of ASB, all those measures are now within the survey data and I think could make it quite useful, just a plug here, a useful piece of analysis if anyone's interested in that sort of data. Alcohol and drug misuse and experience of domestic abuse, sexual assault and abuse during childhood. So it's got a really wide topic range before just producing those headline estimates that we publish quarterly and come out. So thinking about that trend, the reason we've got the trend series is the surveys measured crime in the same way since it was first set up back in 1981. It's a face to face survey, collects details around how you've been a victim of crime by asking people living in private households whether or not they've experienced crime in the last 12 months. It's used a continuous interviewing approach since 2001, but it has evolved over time. I'm not going to linger on this chart because I think Joe's also got a copy in his presentation. I just wanted to point out a couple of the main developments. So we introduced interviewing 10 to 15 year olds. We've also changed to include a measure of fraud and cybercrime, fraud and computer misuse offences, and those have been incorporated into the main estimates. So although the consistency of approach is really important for being able to see that trend line that I showed back here to see that peak in the mid 90s of crime and we can compare back to that, it is important that we make changes as and when they're relevant. So obviously one of the big things that we've had to deal with over the last few years is the same as everyone else is we suspended interviewing in March 2020 on the crime survey for England and Wales due to the COVID pandemic. Temporarily we launched the telephone crime survey, the TCSEW, and that was to provide headline estimates of crime across the pandemic. But we were able to resume face to face interviewing in October 2021. But we are seeing challenges with face to face interviewing now since the return to face to face. I think this isn't something that's unique to the crime surveys affecting all face to face surveys, not just here in the UK but internationally as well, seeing lower response rates and increasing costs, increasingly high costs of data collection using a face to face approach. And largely partly because of that and also requirements for larger sample sizes to be able to better measure sort of geographical, have better estimates at lower geographical levels. We've been looking at ways to increase the sample size and reduce our reliance potentially on face to face interviews. We've also issued, looked at seeing a drop in response rates since the survey returned to face to face fields. So we had quite a nice steady trend line up until really sort of just before the pandemic we started to dip just below 70%. Since we've returned to face to face from October 2021 the response rate has dripped to sort of dipped to the mid 40s. Obviously October to March period, we were still seeing some lockdown measures in place, some restrictions around the COVID pandemic. We are seeing a little bit of an improvement this year, it's gone up to sort of the late 40s I think in the response rate so far. But it is a concern and it's something that we've seen again it's not just affecting the crime survey for England and Wales, it's affecting all face to face surveys. And we are looking at the impact of that on our data and whether or not we're seeing the increased non-response bias in the data. That work is still ongoing because we've only just sort of started to get the full year data and analysis to be able to look at that. So there's a number of reasons for the transformation but partly we've been thinking about how do we transform the survey, how do we make sure it remains fit for purpose. So we want to think about the sustainability of the survey with the increasing cost of face to face. We can't guarantee that we'll be able to fund the survey at the same sort of sample size in future years with those costs going up and up exponentially. So we're looking at a reduced reliance on the face to face interviews and also as I said thinking about that increased precision and granularity of the main estimates of crime. And also modernising the survey with a multimodal approach so it's a bit more respondent focused and people are able to sort of select a mode potentially. So we've introduced a longitudinal design to the survey. We're starting with we've maintained Wave 1 interviews as they are face to face and that's so that we maintain the trend series. We've introduced a telephone survey at Wave 2 so after 12 months after your original interview we're back to responders to ask them again have they experienced crime in the previous 12 months. And then we're back to them at Wave 3 after another 12 months etc. So we have no overlapping reference periods. That was the key thing for us in the design of Wave 2 so we can link people's data together. And then we hope eventually subject to success of the development work that we're undertaking at the moment to be able to introduce an online mode as an alternative to the telephone for Wave 2. Joe will talk more about that when I suspect in a moment. So in all our estimates that we're publishing at the moment we're using Wave 1 data. I've been focused mainly on the implementation of Wave 2 of getting it set up getting the interviews in and Joe is looking at how we then potentially look at combining that data with Wave 1 data and how we maintain the trend series as we move forward with transformation. So it's a brief introduction of transformation before Joe will talk about it shortly. But thinking about the crime survey design for the 22-23 survey I just wanted to update on a few changes that we made since our return to face to face field work. So we adjusted the sample design to allow for a greater flexibility in sample sizes. We saw increasing demand from Home Office in particular for a boosted sample size and to be able to have a little bit more flexibility in the numbers of interviews that we were delivering. So we moved away from the complex sample design that we were using pre-pandemic to a more simple random cluster design which allows a greater flexibility. We've maintained the minimum targets within each police force area and the total sample size for Wave 1's reduced to around 30,000 interviews. We introduced a new module of questions on harassment in April 2022 and we made some small tweaks to those questions at the six month point in October 2023. With the harassment questions we put quite a lot of, we looked back at the data that we had from TCSEW. So we were able to see how respondents responded to a change question within the screen of a TCSEW that actually broadened out the threat screener to cover harassment as well. So we were able to use that data to feed into a more detailed measure of harassment that's gone in as a new module of questions. I think what's important to note about that harassment measure is that we've moved away slightly from the legal definition of harassment as a course of conduct, the repeated nature of the offence. That was to enable us to capture sort of in-street harassment, sort of one-off incidents of harassment that people might be experiencing that might not meet the level that would need, you know, wouldn't count as a legal definition of harassment, but is certainly impacting on people's experience as they're out and about. So just to mention here as well, in the domestic abuse module there was an error in the survey script from October 2022 to January 2023, which means there was missing data for 10 key questions. So the full details of that in the user guide, but if you are using that data, there's a new weight, there's an additional weight on the data file which is, they're just listed here, and those are added to the data file for using that data where you'll need to use the eight months of data. And then in 2023-24 we introduced a high crime area boost sample, so this is particularly to reflect a home office interest in understanding neighbourhood crime in high crime areas. So one of the levelling up missions was to reduce the amount of neighbourhood crime within high crime areas and that's being measured by the crime survey for England and Wales. So we've introduced a boost sample, so that's 6,000 out of the total of 32,500 interviews we're conducting on the survey are taking place in high crime areas. And those are defined using the crime index element of the IMD. We've also launched the redesigned questions on domestic abuse. Those are added as a split sample experiment. We look to update on those in late 2024 and they're planned to be on the survey until March 2025. And currently this year we're running the experience of abuse during childhood module on this year's survey. So, as I mentioned, all estimates for this year are based on wave 1 data. Wave 2 is designed to be implemented with no impact on wave 1 data. And so we're fully exploring the impact of that before we combine anything. And then just a note on children's transformations. So the child crime survey was suspended over the pandemic as well. We interviewed 10 to 15 year olds within households who were selected as part of the main survey. The children's survey wasn't resumed until April 2022. We've seen again a little bit of a drop in the conversion into interviews from the children's survey data. So a drop in the response rates there of 10 to 15 year olds responding, which has made us look to whether there's a better way of interviewing 10 to 15 year olds, not connected with the main adult interview. So essentially that objective there is thinking about how we develop that efficient, sustainable and resilient survey, which delivers a high quality, timely and inclusive data on children. So we've been doing some call and quant research of looking at the feasibility of moving data collection online for the children. And we're looking at a large scale online response rate test launching later this month with a sample of 7,000 children, so sampling from the National Pupil database. And then pilot results will inform a recommendation for the future design of the children's survey with a view to potentially stopping face-to-face data collection for the children and moving fully to online data collection. So I've just put a couple of headlines here from our latest release in January on the crime survey. So overall what we're seeing at the moment is quite a mixed picture with crime. We're seeing some crimes going up in the survey, some going down, some going up compared to last year but not quite a pre-pandemic level. So quite a complicated picture of what's happening to crime at the moment. But overall fraud decreased year on year for 13% of the notable reductions in advanced fee fraud and other fraud as well, which is essentially just fraud that doesn't fall into any of the other buckets. Computer misuse though increased by 30%. And that was mainly driven by that 36% rise in unauthorised access to personal information. And then criminal damage decreased by 21% and again largely driven by 30% for incremental damage to a vehicle. There was no change in CSCW estimates of violence, bear in mind that doesn't include sexual offences except for serious wounding with a sexual motive. But estimates were to still 28% lower than in the year and in March 2020. So I think we've shared these so I've just put the links on to our recent publication so that you can access them from the slides. And then just finally on the availability of data. So the TCSCW non-victim file data sets are now available in the data archive. We don't have any plans to publish the victim file data sets for TCSCW. Largely that's because of the longitudinal nature of those data sets. We didn't have the processes set up to do that. It's a temporary measure obviously to run the TCSCW over the pandemic. So we're focusing more on the archiving now. We've returned to face-to-face of the new face-to-face data sets. So the 22-23 data set is currently in the final stages of preparation. It should be in the data archive by the end of March at the latest. We don't yet have plans for Wave 2 data to be archived. We're focusing on the face-to-face data. That will no doubt be archived in due course but no current plans or timeline as yet. Thank you. Any questions?