Menu Close

The pitfalls of measuring violence against women

Women in Poland stand up against violence. EPA/Bartolomiej Zborowski

It was simultaneously heartening and dispiriting to see the release of a survey by the European Union Agency for Fundamental Rights (FRA) on the prevalence of violence against women in Europe.

Dispiriting because the results shine a light on the continued widespread experience of violence against women that exists, with around a third of all women found to experience physical or sexual violence from the age of 15. We know from our own research in the UK the long-term and significant harm that such experiences can lead to. There is also a risk of us becoming almost de-sensitised to these statistics. We must remember that for each individual involved there may be a back story of human suffering or fear, and a future story of potential harm.

So it was heartening to also see that FRA are using the findings to not only highlight the issue but also make number of recommendations, including calling for EU member states to ratify the Council of Europe’s Convention on preventing and combating violence against women and domestic violence. The authors also recommended that police, healthcare and other professionals need to be trained and properly resourced to identify and reach out to those who have experienced abuse; that specialist support services should be available for those who need it following abuse, and that prevention campaigns should be aimed at men as well as women.

These and their other recommendations should all be welcomed. But the way in which research such as this is reported can also obscure the true issue.

Measuring sexual violence and abuse

A striking finding in the FRA survey that has been widely reported in the media is the difference in reported rates of abuse between European countries. Denmark has the highest level of incidents of sexual or physical abuse at 52%, with the UK fifth highest with 44% and the lowest Poland at 19%.

This immediately raises the questions as to whether this finding really reflects the extent of abuse in these countries or, as is noted in the report, if in some countries it remains less socially acceptable to talk about or recognise certain behaviour as abuse. It also raises broader questions about the challenge of trying to measure levels of abuse.

This area is notoriously difficult. It is well known that many people don’t report abuse or violence to the authorities due to the sense of shame, fear of reprisals or the feeling that the response to their report will not be appropriate. Across different European countries different legal frameworks are also in place meaning that what is recorded as abuse or violence will differ.

In a survey attempting to measure a social phenomenon, definitions are all-important. How you define an issue will affect the number of people who report they have experienced it. There is also what is known as “unacknowledged abuse” when an individual does not perceive themselves to have experienced domestic or sexual abuse – because they have normalised or blocked out the experience – but if it is defined as a situation it is clear that it actually was abusive.

Think for example of the difference of asking someone if they have been sexually assaulted, versus asking if someone had intentionally touched them in a sexual way they did not agree to. Both refer to sexual assault. People who have experienced abuse or violence may not have defined it in such a way, because the way in which it occurs is complex. If you ask someone if they have been raped (forced to have sex) they may be unsure if they have – perpetrators of such violence can use sophisticated tactics such as grooming that lead to the victim feeling they are “to blame” for the abuse.

Both the FRA survey and the Adult Psychiatric Morbidity Survey carried out by NatCen attempt to address this limitation by asking about specific acts of violence, such as whether someone has been “slapped”, “pushed or shoved”, or have been subject to coercive control such as “not being allowed to see friends”, to get to the heart of the matter.

If we think of the recent Savile scandal we know that many more people who have experienced abuse in the past have now come forward – both to the police and also anonymously to survivor organisations, such as calling the national rape crisis helpline. Some of these people have not been able to conceptualise or speak about this abuse for many years, to anyone, until this point. So would they have felt able to note in a questionnaire that they had experienced it?

In some respects then, discussing the prevalence of abuse can be a distraction. That it occurs at all, at any time, in any form, to anyone, is wrong. Much can and should be done to continue the fight to prevent it, and to respond to the needs of those who have experienced it.

Research needs to measure these needs and how we can respond to them. We know it is a complex, difficult issue and we need to have complex, nuanced difficult discussions about it if we truly are to address it.

Want to write?

Write an article and join a growing community of more than 180,900 academics and researchers from 4,919 institutions.

Register now