Online Discussion on proposed indicators for data collection on intimate partner violence, rape and femicide

7 Jul '16 Thu 10:00 CEST07/08/2016 9:00pm EuroGender Online Discussion public Online Discussion on proposed indicators for data collection on intimate partner violence, rape and femicide Europe/Vilnius 07/07/2016 11:00am
8 Jul '16 Fri 20:00 CEST
Past

Node tabs

Discussion
Deirdre Brennan's picture

[quote=Nathalie Meurens]

Regarding the femicide indicator, do you believe it would be relevant to provide comparable age disaggregation (+15, 15-49, 49, other?) and to provide disaggregation for nationality/country of birth?

[/quote]

Hi Nathalie,

Yes I do believe disaggregating by age is crucial, more discussion and understanding is needed on IPV and Child-Parent killings of women over 65. And having the stats to prove this is an issue is a great place to start.

Country of birth, is definitely useful also. I would add to that ethnicity where possible.

Nationality status may be more complicated to collect. 

Dovile Stoskeviciute's picture

As concerns an intimate partner definition, Lithuanian police has welcomed it as to bring certain clarity within domestic violence offenses as currenty, data pertaining domestic violences, does not refer to just intimate partner and includes also family members, even cohabitors, therefore, there is no correct statistics pertainin IPV in Lithuania as such definition is not established by Lithuanian criminal code or other type of applicable laws. However, police in charge of collecting the data also expressed their concern that at times it might be complicated to establish what type of relationship exist between victim and perpetrator if it is not explicitily disclosed and identified. Therefore, submitting such data will depend on assessment of police officer which might not always appear correct as well.

Anonymous's picture
Anonymous (not verified)

[quote=Deirdre Brennan][quote=Nathalie Meurens]

Regarding the femicide indicator, do you believe it would be relevant to provide comparable age disaggregation (+15, 15-49, 49, other?) and to provide disaggregation for nationality/country of birth?

[/quote]

Hi Nathalie,

Yes I do believe disaggregating by age is crucial, more discussion and understanding is needed on IPV and Child-Parent killings of women over 65. And having the stats to prove this is an issue is a great place to start.

Country of birth, is definitely useful also. I would add to that ethnicity where possible.

Nationality status may be more complicated to collect. 

[/quote]

Thanks for your feedback. Indeed, those disagregation would be very useful. In terms of age groups, which age groups would you suggest?

In terms of ethnicity, it is difficult to have EU wide data as a number of countries cannot collect such data by law as they are considered sensitive. 

Council of Europe's picture

The questionnaire developed to monitor the Istanbul Convention might be helpful for this discussion. The type of data it requests from state parties in relation to femicide is:

- the number of deaths of women, the number of such cases in which the authorities اونيلا had prior knowledge of the woman's exposure to violence, the number of perpetrators convicted in relation to these cases and the number of type of sanctions. All of this to be broken down by sex, age, and relationship of the perpetrator to the victim.

- the same information for attempted murders of women, ie cases that didn't lead to the death of a woman.

This covers all of the different forms of femicide without using the term, because the crime and justice sector doesn't operate on this terminology.

You can find the questionnaire on our website

www.coe.int/conventionviolence under GREVIO/evaluations

Federal Office of Justice's picture

[quote=Nathalie Meurens]

Regarding the femicide indicator, do you believe it would be relevant to provide comparable age disaggregation (+15, 15-49, 49, other?) and to provide disaggregation for nationality/country of birth?

[/quote]

First I would like to support the "Femicide Index" as presented by Hana. As Dovile already wrote on behalf of Lithuania, there is no special femicide crime in Germany as well.

Comparable age disaggregation seems to me to be a desireable concept, as long as that kind of data can be obtained from the MS; I even would opt for more disaggregation like 15-20, 21-30, 31-40, 41-50, 51-60, 60+

Disaggregation for nationality/country of birth on the other hand seems to be more problematic, as it might be less likely that this kind of data is already collected. In addition I get the feeling, that there is a presumption of a cultural/religious prevalence at the bottom of this indicator. Therfore the information about the nationality/country of birth for the offender would be as much if not more important to know; and that seems to me to be overloading this indicator.

And if you disaggregate by nationality for femicide, this would be appropriate for the other indicators as well.

Daniela Cherubini's picture

[quote=Hana Spanikova]

It would be useful to create a femicide index that would require sex disaggregating current homicide data, alongside harmonising the breakdown ‘relationship between victim and offender’ and creating coding for ‘female homicide by intimate partner’, what do you think?

[/quote]

I would suggest to further discuss the decision of taking into accunt just femicide in intimate relationships. Or, to further explain and share the reasons that led you to the decision.  I understand the pragmatic rationale of the idea, but I am not sure on the political opportunity to do that. Honour crimes, femicide commited by family members, other crime hates, as well as sexual workers femicides by clients, are also important to detect.

Hana, is your proposal meant to be an alternative way to tackle the issue? To have a femicide index instead of just the indicator of intimate femicide?

Anonymous's picture
Anonymous (not verified)

[quote=Council of Europe]

The questionnaire developed to monitor the Istanbul Convention might be helpful for this discussion. The type of data it requests from state parties in relation to femicide is:

- the number of deaths of women, the number of such cases in which the authorities had prior knowledge of the woman's exposure to violence, the number of perpetrators convicted in relation to these cases and the number of type of sanctions. All of this to be broken down by sex, age, and relationship of the perpetrator to the victim.

- the same information for attempted murders of women, ie cases that didn't lead to the death of a woman.

This covers all of the different forms of femicide without using the term, because the crime and justice sector doesn't operate on this terminology.

You can find the questionnaire on our website

www.coe.int/conventionviolence under GREVIO/evaluations

[/quote]

Thank you for your feedback. We agree on taking on board the disaggregations proposed by the GREVIO questionnaire.

I have a question: would you therefore have attempt murder falling within the scope of femicide rather than IPV?

Council of Europe's picture

[quote=Nathalie Meurens][quote=Council of Europe]

The questionnaire developed to monitor the Istanbul Convention might be helpful for this discussion. The type of data it requests from state parties in relation to femicide is:

- the number of deaths of women, the number of such cases in which the authorities had prior knowledge of the woman's exposure to violence, the number of perpetrators convicted in relation to these cases and the number of type of sanctions. All of this to be broken down by sex, age, and relationship of the perpetrator to the victim.

- the same information for attempted murders of women, ie cases that didn't lead to the death of a woman.

This covers all of the different forms of femicide without using the term, because the crime and justice sector doesn't operate on this terminology.

You can find the questionnaire on our website

www.coe.int/conventionviolence under GREVIO/evaluations

[/quote]

Thank you for your feedback. We agree on taking on board the disaggregations proposed by the GREVIO questionnaire.

I have a question: would you therefore have attempt murder falling within the scope of femicide rather than IPV?

[/quote]

Yes, the choice was between the two but it was decided in favour of linking attempted murder to femicide instead of IPV. For the same reasons as have been raised in this discussion: although some IPV cases might lead to the death of a woman, not all women that are murdered are victims of their partners. It seemed more useful and less restrictive to link it to the issue of women being murdered for reasons of their gender rather than IPV.

Maria Guseppina Muratore's picture

Sorry, I was out for a problem at work.

I 'll try to follow you again 

Dovile Stoskeviciute's picture

Yes I would agree that taking murder within the scose of femicide could lead to certain unification of definitions and would allow to gather more accurate data on such offences.

Elena Fries-Tersch Milieu's picture

[quote=Federal Office of Justice][quote=Nathalie Meurens]

Regarding the femicide indicator, do you believe it would be relevant to provide comparable age disaggregation (+15, 15-49, 49, other?) and to provide disaggregation for nationality/country of birth?

[/quote]

First I would like to support the "Femicide Index" as presented by Hana. As Dovile already wrote on behalf of Lithuania, there is no special femicide crime in Germany as well.

Comparable age disaggregation seems to me to be a desireable concept, as long as that kind of data can be obtained from the MS; I even would opt for more disaggregation like 15-20, 21-30, 31-40, 41-50, 51-60, 60+

Disaggregation for nationality/country of birth on the other hand seems to be more problematic, as it might be less likely that this kind of data is already collected. In addition I get the feeling, that there is a presumption of a cultural/religious prevalence at the bottom of this indicator. Therfore the information about the nationality/country of birth for the offender would be as much if not more important to know; and that seems to me to be overloading this indicator.

And if you disaggregate by nationality for femicide, this would be appropriate for the other indicators as well.

[/quote]

regarding the age groups: yes, normally this should be possible, but we do not have information for all countries whether they collect the data with the precise age OR birth date. We could imagine that police at least collects this kind of data, no? However, for social services, this is unlikely to be the case. They might only use broader categories for recording. 

re. nationality/country of birth: yes, it really does not seem feasible at this stage, although very relevant, especially for harmful practices; furthermore, in some countries admin data is collected on ethnic background and not by nationality, while for other countries it is the other way round. We think that this may be introduced at a later stage as an additional tag. 

Daniela Cherubini's picture

I am sorry, I have to leave the discussion. Thank you very much, I hope my contributions have been of any help. I'll check tomorrow the discussion, answering for direct questions or clarifications.

I'll put here some comments for the last question, which you are going to discuss later: whether the proposed indicators are sufficients to support politcy develpment

Looking at the indicators you are focusing on, we have 1) indicator on IPV, 2) indicator on rape and 3) intimate femicide. This means that we have just one indicator that also cover some forms of violence outside intimate partnership, covering just the more severe form of sexual violence. I see the point, since the IPV is the most common, and the less culturally sanctioned form of GBV. Therefore we need policies that address it. However, the most common does not mean that 99% of gbv is intimate partner violence - there are many experiences and many subjects who will be not included in these indicators. Risk of neglection is high. This could be corrected either transforming the last indicator (indicator of femicide instead of intimate femicide), and/or further expanding this effort and including new indicators int he near future (eg. sexual harrasment, or sexual harrasment at work).

Thank you for your work.

Anonymous's picture
Anonymous (not verified)

[quote=Daniela Cherubini][quote=Hana Spanikova]

It would be useful to create a femicide index that would require sex disaggregating current homicide data, alongside harmonising the breakdown ‘relationship between victim and offender’ and creating coding for ‘female homicide by intimate partner’, what do you think?

[/quote]

I would suggest to further discuss the decision of taking into accunt just femicide in intimate relationships. Or, to further explain and share the reasons that led you to the decision.  I understand the pragmatic rationale of the idea, but I am not sure on the political opportunity to do that. Honour crimes, femicide commited by family members, other crime hates, as well as sexual workers femicides by clients, are also important to detect.

Hana, is your proposal meant to be an alternative way to tackle the issue? To have a femicide index instead of just the indicator of intimate femicide?

[/quote]

Thanks for raising this. Yes, ideally we would have a very encompassing scope for femicide. However, our goal was to have an indicator which could be populated by existing data. We made the choice to focus on forms for which data was sufficiently available as a starting point. 

Tugce Tugran's picture

[quote=Daniela Cherubini]

I am sorry, I have to leave the discussion. Thank you very much, I hope my contribution have been of any help. I'll check tomorrow the discussion, answering for direct questions or clarifications.

I'll put here some comments for the last question, which you are going to discuss later: whether the proposed indicators are sufficients to support politcy develpment

Looking at the indicators you are focusing on, we have 1) indicator on IPV, 2) indicator on rape and 3) intimate femicide. This means that we have just one indicator that also cover some forms of violence outside intimate partnership, covering just the more severe form of sexual violence. I see the point, since the IPV is the most common, and the less culturally sanctioned form of GBV. Therefore we need policies that address it. However, the most common does not mean that 99% of gbv is intimate partner violence - there are many experiences and many subjects who will be not included in these indicators. Risk of neglection is high. This could be corrected either transforming the last indicator (indicator of femicide instead of intimate femicide), and/or further expanding this effort and including new indicators int he near future (eg. sexual harrasment, or sexual harrasment at work).

Thank you for your work.

[/quote]

Thank you very much Daniela for being active and useful contributions. Looking forward to collaborate further in the future. 

Anonymous's picture
Anonymous (not verified)

In your opinion, what challenges are likely to arise if this femicide indicator is implemented at national level? How can these be overcome?

Daniela Cherubini's picture

[quote=Tugce Tugran][quote=Daniela Cherubini]

I am sorry, I have to leave the discussion. Thank you very much, I hope my contribution have been of any help. I'll check tomorrow the discussion, answering for direct questions or clarifications.

I'll put here some comments for the last question, which you are going to discuss later: whether the proposed indicators are sufficients to support politcy develpment

Looking at the indicators you are focusing on, we have 1) indicator on IPV, 2) indicator on rape and 3) intimate femicide. This means that we have just one indicator that also cover some forms of violence outside intimate partnership, covering just the more severe form of sexual violence. I see the point, since the IPV is the most common, and the less culturally sanctioned form of GBV. Therefore we need policies that address it. However, the most common does not mean that 99% of gbv is intimate partner violence - there are many experiences and many subjects who will be not included in these indicators. Risk of neglection is high. This could be corrected either transforming the last indicator (indicator of femicide instead of intimate femicide), and/or further expanding this effort and including new indicators int he near future (eg. sexual harrasment, or sexual harrasment at work).

Thank you for your work.

[/quote]

Thank you very much Daniela for being active and useful contributions. Looking forward to collaborate further in the future. 

[/quote]

thanks to you and all the participants

Elena Fries-Tersch Milieu's picture

[quote=Dovile Stoskeviciute]

As concerns an intimate partner definition, Lithuanian police has welcomed it as to bring certain clarity within domestic violence offenses as currenty, data pertaining domestic violences, does not refer to just intimate partner and includes also family members, even cohabitors, therefore, there is no correct statistics pertainin IPV in Lithuania as such definition is not established by Lithuanian criminal code or other type of applicable laws. However, police in charge of collecting the data also expressed their concern that at times it might be complicated to establish what type of relationship exist between victim and perpetrator if it is not explicitily disclosed and identified. Therefore, submitting such data will depend on assessment of police officer which might not always appear correct as well.

[/quote]

alright, but regarding the identification of victim-partner-relationship by the police, are there any possibilities of improvements? how could it be improved?

Jurgita Peciuriene's picture

[quote=Elena Fries-Tersch Milieu][quote=Dovile Stoskeviciute]

As concerns an intimate partner definition, Lithuanian police has welcomed it as to bring certain clarity within domestic violence offenses as currenty, data pertaining domestic violences, does not refer to just intimate partner and includes also family members, even cohabitors, therefore, there is no correct statistics pertainin IPV in Lithuania as such definition is not established by Lithuanian criminal code or other type of applicable laws. However, police in charge of collecting the data also expressed their concern that at times it might be complicated to establish what type of relationship exist between victim and perpetrator if it is not explicitily disclosed and identified. Therefore, submitting such data will depend on assessment of police officer which might not always appear correct as well.

[/quote]

alright, but regarding the identification of victim-partner-relationship by the police, are there any possibilities of improvements? how could it be improved?

[/quote]

And what are the main challenges?

Elena Fries-Tersch Milieu's picture

If there are no more comments on femicide we would like to raise a general question related to all indicators: 

Do you think that the proposed indicators are sufficient to support policy development?

Andrada Filip's picture

[quote=Elena Fries-Tersch Milieu]

We would like to ask Andrada from the WAVE network specifically: do you think the indicators on IPV (the general one and possibly also the specific ones) could be populated with social services data? How do you see the potential of social services data for this? Our research showed that quite a few Member States collect data from social services nation-wide, but there is of course the problem that they have different types of social services and different target groups. Do you have any suggestions for adaptation of the indicator for social services' data? 

THANK YOU!

[/quote]

Apologies for the late reply but I have been caught up with work at the office. In any case I have consulted my colleagues, and this would be our suggestions: 

For the general indicator, we envision the following challenges and some ways to mitigate them:

1.       Social services would have to be defined more narrowly and I imagine a coordinating body would have to be in charge of collecting these data from a pre-defined number of social services. Important to consider would be whether general services such as legal aid offices or employment agencies, or specialist services like national women’s helplines, shelters.

2.       One should be careful with the “as a share of the total population of women” because it may lead to troublesome interpretation, when comparison between countries is attempted. If a smaller percent accesses services in Romania as opposed to in another country, it may be indicative of a lack of awareness about services, the times the services are available (e.g. 24/7, not 24/7), availability of staff, among others. A helpline in a country with similar population may have significantly different number of callers depending on staffing and opening times. Furthermore, shelter capacity differs per country, in terms of the number of beds or family places available, the stay allowance (e.g. some shelters allow for up to 12 months, while others only 3 months), which will all impact the number of victims that are supported, and distort the ability for clean comparison.

3.       The method of service provision may impact numbers. For example, some helplines may provide telephone service and also chat or advice via e-mail. Some centres conduct outreach, whereas others require victims to physically visit the establishment.

4.       For a helpline, which is by nature and according to international standards required to be confidential, it is not likely that data on age will be collected.

5.       Specialist women’s services are not always funded in a stable fashion to make sure they operate continuously. Not only can the hours of operation for a helpline change from one year to the next, shelters and centres can also close year to year, and new ones can open. In some cases, in one year, within a network of women’s shelters in any given country, the number of shelters feeding national level data may change.

6.       Always keep in mind that services deal with repeat clients, but due to confidentiality this may not always be recorded.

7.       It will be important to allow for quantitative data to be supported with some qualitative feedback to provide the context within which service provision exists, to allow for some preliminary explanation for the data.  

The number of incidents reported to anyone varies due to cultural differences – e.g. in some countries it is easier to admit to IPV than in others, this distorts the picture of “what is really going on”.  And differences in legal frameworks can also impact e.g. some countries have Emergency Barring Orders which allow some women to deal with IPV in a different way and how the police use the legislation on Emergency Barring Orders varies enormously (see page 86 of WAVE 2015 Report).  This could impact on how many women approach a shelter for accommodation, how could this variation be allowed for.  Finally, the amount of resources a woman has impacts on how she can cope with IPV e.g. if she is able to rent private accommodation and pay for a therapist, she may never approach any social services for help.  Similarly, if she has an extensive support network she may not approach any social services.  Just counting who approaches services shifts the focus of the picture towards women with fewer resources (poorer women & possibly migrant women).

Maria Guseppina Muratore's picture

About femicide, I think that the only indicator useful it is femicide that is a very complex concept to be definied and operazionalized from a  statistical point of view, but the female homicide by victim/perpetrator relationship.

Recently in Vienna (UNODC), we had a meeting on this issue and what emerged is the difficulty to transalte femicide in data.

What it is possibile is to identify women killed by a partner/ former partner, as well by other family members, or the homicide of prostitutes or of the honour killing and so on.

The clear problem was that very few countries had statistics on this topic and also for latin america countries, that has the law agasint femicide, data are not easy to be collected

at the end of the meeting, the decision was to start collecting the victim/perpetrato relationship.

About age, I agree about the classes of age comparison, starting from before 18 years old.   

Deirdre Brennan's picture

[quote=Nathalie Meurens][quote=Deirdre Brennan][quote=Nathalie Meurens]

Regarding the femicide indicator, do you believe it would be relevant to provide comparable age disaggregation (+15, 15-49, 49, other?) and to provide disaggregation for nationality/country of birth?

[/quote]

Hi Nathalie,

Yes I do believe disaggregating by age is crucial, more discussion and understanding is needed on IPV and Child-Parent killings of women over 65. And having the stats to prove this is an issue is a great place to start.

Country of birth, is definitely useful also. I would add to that ethnicity where possible.

Nationality status may be more complicated to collect. 

[/quote]

Thanks for your feedback. Indeed, those disagregation would be very useful. In terms of age groups, which age groups would you suggest?

In terms of ethnicity, it is difficult to have EU wide data as a number of countries cannot collect such data by law as they are considered sensitive. 

[/quote]

If it helps the age groups we use are as follows:

Under 18

18-25

26-35

36-45

46-55

56-65

66+

If there is scope then it is also worth identifying the age band of the perpetrator which the police will also have on file.

Dovile Stoskeviciute's picture

Currently when gathering data on committed offences, Lithuanian police mainly identifies if there are any close family relations between victim and perpetrator. In respect to domestic violence offfences, it can be also identified if two people of cohabitants. The process could improve if an intimate partner definition would be introduced and established on the national level in Lithuanian as well, as for a moment, we officially dont have even a civil partnership as a legal status between two cohabitting people and the only indicator which could be actually identified pertaining these offenses is if victim-perpetrator are spouses. 

Barbora Holubová's picture

[quote=Elena Fries-Tersch Milieu]

If there are no more comments on femicide we would like to raise a general question related to all indicators: 

Do you think that the proposed indicators are sufficient to support policy development?

[/quote]

I think that a great job was done and many issues were /or will be taken into the consideration, including  some suggestions from this discussion. Despite of the limits, this would  be definitelly a progress.

Good-buy and thank you for the opportunity for discussion.

Barbora Holubova

Tugce Tugran's picture

[quote=Maria Guseppina Muratore]

About femicide, I think that the only indicator useful it is femicide that is a very complex concept to be definied and operazionalized from a  statistical point of view, but the female homicide by victim/perpetrator relationship.

Recently in Vienna (UNODC), we had a meeting on this issue and what emerged is the difficulty to transalte femicide in data.

What it is possibile is to identify women killed by a partner/ former partner, as well by other family members, or the homicide of prostitutes or of the honour killing and so on.

The clear problem was that very few countries had statistics on this topic and also for latin america countries, that has the law agasint femicide, data are not easy to be collected

at the end of the meeting, the decision was to start collecting the victim/perpetrato relationship.

About age, I agree about the classes of age comparison, starting from before 18 years old.   

[/quote]

Thank you very much that is quite useful actually. We will take that into consideration.

Tugce Tugran's picture

[quote=Inger Lövkrona][quote=Tugce Tugran][quote=Inger Lövkrona][quote=Elena Fries-Tersch Milieu]

 

If there are no more comments on femicide we would like to raise a general question related to all indicators: 

Do you think that the proposed indicators are sufficient to support policy development?

[/quote]

As to develop policies to collect information on GBV in order to get a picture of prevalence and forms of GBV in a country, the indicators could be supportive. However, as has been discussed, some clarifications regarding the contexts of sexual assault, rape and femicide are necessary. I think it also would be worth considering, as proposed, a definition of the concept "intimate partner". فوائد فيتامين سي للوجه It is also important to cover IPV cases with other family perpertrators.

I also find the indicators supportive to policies on planning for health care, social service support, e g the societies support to the victims.

As to policies to erase, stop, men´s violences, which is the overarching goal för EU:s work on GBV, the indicators are less usefull. To be frank, you cannot combat violence with numbers and definitions. In order to stop the violence you have to address  gender-power structures - and the perpertrators, e g the men.

Gender equality achievements will empower women, however there is no reearch that confirms that gender equality will stop men´s violence.  Sweden is an example of this - we have a very high ranking on the global gender equality scale, but the vilence against women is not decreasing.

[/quote]

Dear Inger, I agree with you, of course the problem is much wider and deeply anchored in societal structures, power relations between victims and perpetrators. The numbers, data is a way among others to fight this phenomenon that is unfortunately common in every country, every culture. اونيلا We are hoping to advance things incrementally on every front.Hence this study.  About the issue of equality and violence, one can also argue that in Sweden the incidence is higher not because there is more violence against women but Swedish women are less afraid to report incidences, and there is a higher awareness about the issue. We can only hope that policies aiming to reduce inequality between men and women will also contribute to fight the violence.   quote=Inger Lövkrona]

[/quote]

I sincerly hope you are right, at least gender equality has encouraged women to report, but unfortunately there are too many cases to report, still...

I´ll leave you now, thanks for an interesting debate

[/quote]

Thank you Inger for the input and insights! 

 

Siobán O’Brien Green's picture

The indicator (Femicide) could be measured by the following sources and units:

1.    Source: police records of crimes; units: number of women victims.

2.    Source: hospital and healthcare professionals’ records, units: number of women victims; death records of women

Hello apologies for my late replies to the discussion, I'm Siobán based in Dublin, Ireland. I can't find the section that relates to the sources for Femicid data collection, see above.

I would propse that information and data from coroners offices and annual reports is used here, it is likely that health and hospital records will not rovded the units we are seeking. 

Anonymous (not verified)

Dear Sioban,
Many thanks for your comment. You can find the indicators in the background note, which you can download from the introductory text above which includes the agenda.

Maria José Carrilho's picture

Examples of indicators on intimate partner violence:

  • Number of reported cases of women who experienced physical and/ or sexual/ psicological violence by their current or former partner;
  • Proportion of women who experienced physical and/ or sexual/ psicological violence by their current or former partner in relation to the total number of women resident in the country (per 100 000).
Henriette Jansen's picture

Further to the discussion on definition of partner. I found the following in the 2015 CDC manual on intimate partner violence surveillance.

Intimate partner: An intimate partner is a person with whom one has a close personal relationship that may be characterized by the partners’ emotional connectedness, regular contact, ongoing physical contact and sexual behaviour, identity as a couple, and familiarity and knowledge about each other’s lives. The relationship need not involve all of these dimensions.

Intimate partner relationships include current or former:

  • Spouses (married spouses, common-law spouses, civil union spouses, domestic partners)
  • Boyfriends/girlfriends
  • Dating partners
  • Ongoing sexual partners.

Intimate partners may or may not be cohabiting. Intimate partners can be opposite or same sex. If the victim and the perpetrator have a child in common and a previous relationship but no current relationship then by definition they fit into the category of former intimate partners (CDC 2015).

In my own work countries differ as to what constitutes common-law marriage and even as to whether they would consider anyone who is not a current or former spouse a partner.

Anonymous (not verified)

Dear Henriette, thank you for the information on the definition of intimate partner. It is an exhaustive one. The Istanbul Convention only refers to partners as defined by national laws. So our approach is to compare national definitions and find the most common elements (such as current and former spouses, cohabiting partners, civil partners and dating partners), taking into account the ICCS (which also include current and former spouse, cohabiting partner, non-cohabiting partners).

Elena Fries-Tersch Milieu's picture

We had an internal discussion about your comments on the use of the indicators and the reference population. We came to the conclusion that the figures from administrative sources (so, figures on reported cases) will serve to inform monitoring of programmes and policies that target these services. The figures on reported cases will need to be used in context with other information, and depending on the approach of the monitoring, total numbers or rates will need to be used for cross-country comparison. 

For example: monitoring could compare the effect of a policy (e.g. improvement of accessibility) by comparing TRENDS across countries ('over the period of XY years, country A implemented a programme and in country A reported cases increased while country B did not implement a programme and reported cases stayed the same' - in this case, there is no need for a rate. However, monitoring could also compare effects horizontally: 'country A and country B spend the same budget on police intervention against IPV, but in country A reporting is much lower than in country B, although the prevalence rates are the same' - in this case, total numbers cannot be compared and they should be divided by a reference population that somehow reflects the size of the target group - which would be ever-partnered women; however this information will be hard to get, so the proxy reference group could be all women. 

Do you agree with this explanation and that indicators will be suggested both in total numbers and as rates? 

thank you!