How does crime severity predict victim willingness to meet the offender?

The puzzle

Imagine that you have been a victim of a crime. Afterwards, the police put you in touch with an organisation offering to arrange a meeting between you and the person who committed the crime. Are you more likely to accept this meeting if the crime was minor, or if it was serious?

Studies with members of the public show that most people think they would be more likely to accept the meeting if the crime was minor. Yet studies with people who have actually been victims show that this may not be the case. The relationship between the severity of the crime and victim willingness to meet the offender varies from one study to another, sometimes it is negative, but sometimes it is absent or even positive, i.e., some studies found that victims of serious crimes were more willing to meet the offender than victims of minor crimes.

Solving the puzzle? Background to the current study

In 2015 I added to the confusing set of existing studies, by contributing some confusing findings of my own! I evaluated whether victims of young offenders were more likely to want to meet when the crime was minor, using data from a Youth Offending Team in the UK. You can see the original article here, and a short summary here, but the findings are complicated by the multiple options victims were offered as types of ‘restorative justice’. My tentative finding that victims of ‘medium-seriousness’ offences appeared most likely to meet the offender suggested that the relationship between seriousness and willingness to meet might be a curve (with a peak in the middle, like an n-shape or an upside-down ‘u’). However, this was not possible to confirm with the Youth Offending Team data, so I had to look elsewhere.

One option was to try another study in a ‘naturalistic’ setting like the one at the Youth Offending Team. But the problem with this type of study is that it is not possible to keep consistent many factors that might affect victims’ decisions, such as when, how and by whom they are offered a meeting with the offender. Also, many potential participants were effectively filtered out by the courts, the referrers or the facilitators – leading to all sorts of selection biases. At the other end of the spectrum, another option would be to investigate the relationship by giving random participants an entirely hypothetical crime scenario and the hypothetical option of meeting the offender, and asking them whether they would meet the offender. However, this extremely ‘unnatural’ setting would not capture real victims’ views, and would be unlikely to capture the types of emotions people really feel after experiencing a crime.

I found a good middle-ground between an entirely naturalistic and an entirely unnatural setting, in the data from the Crime Survey for England and Wales. This survey asks a very large number of people who have been victims of real crimes whether they would have wanted to meet the offender had they been given the opportunity. The victims and the crimes are real, but the offer of a meeting is hypothetical. This means the wording and explanation of restorative justice was consistent across participants, the time since the offence was random (up to 12 months), and all victims were asked the question regardless of the crime they suffered.

Putting the pieces together: Findings

In summary, I found no relationship between overall ‘severity’ of the crime and the victim’s willingness to meet the offender. Severity was measured in the survey by asking the victim to rate how serious they thought the crime was using a scale from 1-20. And victims at all points on the scale were similarly likely to say that they would have liked to meet the offender if they had been given the opportunity (around 25%). I explored my pet theory that the relationship might be a curve, but there was no evidence of that either.

Interestingly, however, that was not the end of the story. Two other factors - related to overall severity - did affect how likely the victim was to be willing to meet the offender: the impact the offence had on the victim’s life, and whether the victim described the offence as violent.

• Impact: The impact of the offence had a positive relationship with victim willingness to meet the offender. In other words, on average, those who were most affected by the crime were most willing to meet the offender.
• Violence: Whether the victim considered the offence to be violent or not[1], had a negative relationship with victim willingness to meet the offender. In other words, victims of crimes that were violent were less willing to meet the offender.

Both relationships were statistically significant while controlling for the other, and for lots of other variables such as age, gender, and ethnicity of the victim.

Often, these two variables are related to one another; in many cases the impact of a violent offence is greater than a nonviolent offence. For most offences, therefore, the positive and negative effects of impact and violence effectively cancel each other out; victims of violent, high-impact offences are similarly likely to want to meet the offender as victims of nonviolent, low-impact offences.

However, violence and impact are not always proportional to one another, so we can see the independent effects of violence and impact when one is low and the other high. Victims of nonviolent offences that are high in impact, for example, are most likely to want to meet the offender. This might explain the anecdotally high number of families of victims of death by reckless driving who choose to meet the offender. The impact of the crime is astronomically high, but the offence was not an intentional act of violence. Conversely, victims who are not personally affected even when the crime is violent, are least likely to want to meet the offender.[2]

Conclusion

We must stop assuming that victims of serious offences will not want to meet the people who committed a crime against them. Of course, in certain circumstances there may be legitimate reasons not to offer restorative justice interventions, but if criminal justice professionals hesitate to make the offer just on the basis that the victim ‘probably wouldn’t want to anyway’, they are unjustifiably denying those victims (and their offenders) access to a potentially valuable service.

The independent effects of impact and violence may explain why people who are not victims tend to think that victims of more serious crimes would be less likely to meet the offender – because they consider only violence, and forget impact. When you considered the hypothetical question at the beginning of this blog, what did you think I meant by ‘serious’? You may have imagined a serious crime to be a violent one, and a minor crime to be less violent. If you did not think about the impact on the victim and used only violence as your indicator of severity, then you would have been right to answer that victims of more ‘serious’ crimes would be less likely to meet the offender. And you would be excused for not thinking about the impact of the crime on the victim, as it varies widely, and not always in proportion to other measures of seriousness. It is impossible to estimate impact in a hypothetical scenario, and even with details of real crimes the impact is hard to gauge unless you hear directly from the victims themselves.

[1] The study unfortunately only used a binary measure, asking victims to describe whether the crime they experienced included any threat or use of ‘violence’ (y/n).

[2] I also investigated whether there was a statistical interaction between these two variables, but there was no evidence of an interaction in this study.

So, are there any rules we could apply to predict which victims are most likely to take up the offer of restorative justice? On the basis of these findings, I argue that the nature of the crime should not be used to determine who is offered the chance to meet the offender. The best predictor of participation was the impact the crime had on victims’ lives, and only the victim can tell us about the size of that impact. The only rule we should apply, therefore, is that we won’t know whether a victim wants to meet the offender until we ask.

Postscript: Offence ‘type’ and other predictors

You may be reading this and wondering whether victims’ choices are more affected by the ‘type’ of crime than the severity. For example, surely victims of sexual offences are less likely to want to meet the offender than victims of other types of crimes? I included this as a control variable, and yes, fewer victims who said the crime included a ‘sexual element’ would have been willing to meet the offender (around 20%) than the rest of the sample (25%). However, when controlling for the other variables this was not a significant predictor of their choice to meet the offender – in other words, both the impact of the crime and whether the victim considered it to be a violent crime, were better and more important predictors of the victim’s choice to meet the offender, compared to whether the offence was of a sexual nature.

Some of the other variables I included as controls were significant predictors of willingness to meet the offender. Victims tended to say they were more willing to meet the offender if any of the following were true: if the victim thought the offence was racially motivated (although there were comparatively few of these cases in the sample), if the offence was committed by a stranger, if the police were informed, if there was contact of some kind between victim and offender during the offence, if the person had been victimised more than once in the last 12 months, or if the person who was victimised was older or identified as Black (compared to the aggregated Asian, White, and ‘Other’ categories determined by the survey). These findings suggest that victim characteristics, the nature of the relationship between victim and offender, and their view of the criminal justice system, may be more important predictors of participation than the nature of the crime.