 OK, next up we have Eloise Davies, so Eloise is an early career researcher in criminology and she's currently working as a research associate at Lancaster University as part of the Violet's Health and Society consortium. Hi, so I'm going to just give a quick overview of one of the pieces of work that I've been working on as part of the vision consortium so it's looking at violent crimes with more than one perpetrator. So when there's more than one assailant understanding the variation of victim's needs. Oh, yeah, it works now. So just to acknowledge our funder. So a quick background to this work, so in the crime survey for England and Wales between 2011-12 and 2019-20 more than a quarter of violent crime was perpetrated by multiple perpetrators. So where there was more than one perpetrator involved in one event of violence. So past research on crime by multiple perpetrators looks at mainly the perpetrators and not the victims. So focusing on the group composition, so the age of the group, the gender of the group and the number of perpetrators in the group. And they find that generally co-offending groups are more often same sex groups of men and they're more often younger than solo offenders. When looking at violent crime specifically there's a discussion about quantity and quality of the violence. So quantity refers to the number of violent crimes that groups of offenders commit together. And quality looks at the severity of the violence. And this research has shown that when there's more than one offender both the quantity and the quality of violence can expect to increase. So multiple perpetrator violent events, mostly on the slide there's MPVs are more often severe and more often result in injury to the victim. And this is compared to single perpetrator violent events. So given the greater severity of crimes perpetrated by multiple perpetrators we might expect that victims more often seek help from various services. And that's what I'm looking at in this paper. So previous research on co-offending usually relies on administrative data. So data from police, health services and specialist services. And one of the issues is this is that you only have data on the victims that come to the service. So I'm looking at using the crime survey for England and Wales to compare both groups of victims and just get a more comprehensive picture of which victims contact which services. So the paper at the slide say eight years, the paper actually uses nine years of crime survey data from 2011-12 to 2019-20. And I use only victim forms related to violent crime, defining violent crime as common assault, serious wounding, attempted assault, other wounding, sexual offences, robbery and attempted robbery. So I don't include threats just because I wanted to look at victims that go to the hospital for treatment so it made sense to limit it to physical violence only. So this gave 6,838 victim forms that when weighted to the population was 7.65 million victim forms or violent events. And there were slightly more victim forms for male victims than female victims so it's 52% male and 48% female. And 73% of the violent events were single perpetrator violent events and 27% were multiple perpetrator violent events. And of the events of multiple perpetrators, 40% had two perpetrators, 20% had three perpetrators and 40% had four or more perpetrators. So this paper has two main research questions. So what are the characteristics of victims of multiple perpetrator violent events and how does this compare to victims of single perpetrator violent events? And are victims of multiple perpetrator violent events more or less likely than victims of single perpetrator violent events to access different forms of services? So the characteristics I looked at were limited to age, sex, ethnicity of the victim and I also looked at victim-perpetrator relationship. And then the services that I looked at were victim support which is a national charity in England and Wales that provides support to victims. It's government funded and led by volunteers and they provide all sorts of services so help with accessing other organisations, help with contacting places to get compensation and other forms of support. Is that from? So the services that I looked at were victim support like I said, whether the victim reported, whether the police were made aware of the incident and whether the victim was injured and whether they ended up getting treatment in hospital. So the first table is probably quite small because I know the fonts have been a little bit. So these show the victim characteristics and compared to single perpetrator violent events and multiple perpetrator violent events. So I'll start at the top and work down. Where there's an asterisk just means that the CHI square test was significant so there's a significant difference between the two groups. So for sex of the victim of single perpetrator violent events these were usually fairly even male and female with slightly more male victims than female victims. But for multiple perpetrator violent events these were majority male victims so 71%, 72% male victims and 28% female victims. And then for ethnic group it was less likely for the victim to be white so there were 13% from minority ethnic groups compared to 9% for the single perpetrator violent event group. And then for age the age of the victim was generally younger than the age for the single perpetrator violent events and then for relationship to the victim. It was quite a bit less common for the victim to know any of the perpetrators if there was multiple perpetrators involved. Especially if the perpetrator was domestic relations so there was less than 3% of multiple perpetrators that were domestic relations to the victim. So I used binary logistic regressions looking at the four types of contact with services and mainly focusing on whether there was multiple perpetrators or single perpetrators. Each logistic regression model controlled for socio-demographics of the victim, relationships of the perpetrator, sex and age of the perpetrator. I didn't include ethnicity of the perpetrator because there was quite a lot of missing in this variable with it being victim reported and socio-economic class of the victim. So for model 1 whether the victim contacted victim support this wasn't significant whether there was a multiple perpetrator or not. For whether the police were made aware of the event there was a 24% increase for when there was multiple perpetrators compared to single perpetrators. For whether the victim was injured, there was a 33% increase in the odds of injury for victims of multiple perpetrator violent events compared to single perpetrator violent events. For whether the victim had treatment in hospital the odds of treatment in hospital were double for victims of multiple perpetrator violent events, than for victims of single perpetrator violent events. I might have rushed through this a little bit too quickly. A bit of a round-up of the main conclusions of this paper. As most research on multiple perpetrators of violent events focuses on the perpetrators, the group composition and the dynamics between the group, they tend to use administrative data sources. So, like I said before, health police records and special services records. This paper that uses the crime survey data shows that victims of multiple perpetrators are more likely to be injured, to report to the police, to turn up at the hospital. So, using the administrative data, you might find that you are... Sorry, I'll try and word that better. So, using the administrative data, obviously there's only data on the people that reach the services. So, the findings of this paper show that there might be an underrepresentation of victims of single perpetrator violent events in administrative data. So, if we're using this data, we should just be aware that there are limits to the coverage. Obviously, everybody here or most people here will be using the crime survey or a form of crime survey. So, it's just an acknowledgement of how valuable this data is and how much it can show us more of a whole picture than other data sources, which allows us to make these comparisons. That's... I finished really early, didn't I?