Civil Protection Order Enforcement

Bernie Auchter: Thank you. I'm Bernie Auchter with the National Institute of Justice — glad to be here. And welcome all of you from your many different organizations. I want to thank the National Council as well as OVW for hosting this — it's very much appreciated — and to Barbara for getting it out beyond this locale and also to folks in Reno.

NIJ and OVW have done many things collaboratively over the years, and this, to me, is certainly an excellent example of our collaboration, and there will be many more of them. At the local level, like in Lexington, Ky., there's a great collaboration going on with Teri [Faragher] and TK [Logan] — and we'll hear much more about that later.

Today, TK is going to present for about 40 minutes, and then Teri will have about a 10- or 15-minute follow-up. Then we'll have a half hour of discussion and Q&A as well. Hopefully, the discussion will get into, “OK, what do we do with all of these research results, and how do we apply them?”

With that, Teri Faragher is the executive director of the Domestic Violence Prevention Board in Lexington, and TK Logan is a professor at the University of Kentucky and has been for a number of years. She's also an author of many scholarly articles on stalking and sexual assault, and a number of books as well. Their more detailed bios are in one of the handouts that are located outside on the table.

With that, I'd like to ask TK to get us started.

TK Logan: Well, first of all I want to say thank you so much for having us here and for the opportunity to share some of the findings that we have from the Kentucky civil protective order study. I want to start out with a little bit about where my motivation for this particular study came.

I was doing another study and using women with protective orders as sort of a sampling frame, and I was studying health, mental health, substance abuse, and how that changed over time with victimization — and other kinds of victimizations. I didn't know much about protective orders at the time — I thought I did — but I was in for some “schooling,” and I realized that I didn't really know that much about it. And also, there was a huge disconnect, for me. We studied, in that particular study, more than 750 women from multiple jurisdictions — rural and urban — and we did a 12-month follow-up. What we found in that study is that 40 percent of the women reported there were no protective order violations. Now this was not a protective order evaluation study, but at the same time, I was hearing from judges all the time, “Oh, women just come back and drop the order …” I found 9 percent in that sample dropped that order in that 13-month post protective order. Women come back; they're always changing it. I found about 20 percent who went back for amendments, and there were a few who made back-and-forth kind of amendments. Victims' advocates were telling me, “It's just a piece of paper, it doesn't work …” And then you look to the media—there are always these stories in the media of the failure of a protective order.

So, these things were not making sense to me. I had women, the survivors, telling me some difficult stories. Some of them talking about the barriers to get the protective orders they had to overcome; talking about some nightmare stories on enforcement. But overall, all the women had better lives after the protective order, so it just wasn't fitting for me. And that's where the motivation for this study came — just from some questions.

The first question is: How effective are protective orders?

I went to the research and guess what I found in the research? Anywhere from 20 to 70 percent of the protective orders are violated. I personally didn't find that very helpful, actually, and of course, there are methodological differences, so if you look at it from official data, it's going to be much lower than if you talk to survivors. And also that question — how effective are protective orders?—is actually a little more complex than just looking at violations, and I really couldn't find one study that gave us the bigger picture.

Second, if protective orders don't work for some women, what are those circumstances? What are those factors associated with violations?

Here's my big one — and I'm not saying I have answers to this — but if protective orders work, then how can we talk to the justice system so they can do more of the good stuff and improve on the areas of weakness. I thought and thought and thought — because I had presented some findings from the earlier study, and they thought it was interesting, but it didn't seem that much had changed.

So that's where the idea for what it costs … What we had done in the past — I had worked with an economist — and we did a study on outcomes of drug court. I really saw a huge transformation when we looked at the costs and the avoided costs to society with that, all over the state, including the justice system as well as legislatures. So I thought, “Why can't we apply that to protective orders?” And I wrote that in, as a big piece.

And finally, like I told you, we were looking at rural and urban women in that first study and I felt like there were differences, but there was nothing I could put my finger on. It was difficult to measure, I couldn't find anything number-wise. And then when I went to the literature on rural experiences of partner violence, they would say, “Well, we found these trends … this is why it's different …” But I would be like, “But we saw some of that with our urban women as well.”

So, I really wanted to do a much better job at comparing rural and urban — not just victims' voices — but also trying to look at community context, so we also did key-informant interviews, we used secondary data and other kinds of research that we could get our hands on. The rural area that we looked at in this study is called the Appalachian rural area. It's a unique cultural tradition, so there's a lot of writing on that, and so that was really nice to integrate to help us interpret our data as well.

But first, of course, nobody else would have funded a study like this so I have to thank the National Institute of Justice. [Audience applauds.] Bernie Auchter was also really supportive and helpful through the whole process. Of course there wouldn't have been a study unless the survivors were willing to talk to us and tell us their stories. And to sit with us — the baselines were three or four hours — we did a three-month and a sixth-month follow-up. At six months, we had a 99 percent follow-up rate, so they stayed in touch with us, and we were able to get their stories.

We also worked with community advisory boards at the initiation of the study and throughout the study. But there were others also in the community that made the study possible, that weren't necessarily on our community advisory board. For example, judges, who allowed us as guests into their courtrooms; that's where we recruited women. And of course my co-authors on this study — Teri Faragher, who's here; Robert Walker; and William Hoyt, who's actually the economist who analyzed and did all the “voodoo” on the cost data.

Just to give you a little background, in the write up, I wrote it as three sub-studies. But in reality, it didn't work like that, because the study is so huge, with so many different components, that I did write it up as three sub-studies, with the first one being the contextual community differences.
 
We did more than 200 key informant interviews; we ended up with a final sample of 188. These were hour-long interviews with professionals; they gave us an hour on the phone to tell us how the process worked in their communities, what they thought the weaknesses were, the barriers and those types of things. We used official court data — you're going to see some of that—we used other research, literature and news articles.

Another thing about our Appalachian rural area is that it's actually gotten national media attention because of the drug use — the drugs have just torn those communities apart.

The second sub-study that we talk about in the report is the interviews with the petitioners who were recruited out of court, and we also got secondary court data that matched those cases so we could look at that.

And the third sub-study was the cost component, which sounds really easy, but it was gathered from the interviews, as well as going out and attaching a cost to a specific service use — and that kind of thing is a lot of groveling for data, which I'm willing to do, obviously. I found a great state source — I got most of our costs that are state-specific; there were a few national data sources that I had to use because we didn't have state-specific data.

Just to give you a little background. Before I get started on the effectiveness, I want to give you a little picture on what the population looked like. On average, women were about 33 years old; 80 percent had children — the vast majority of those children were under 18 years old. Twenty-seven percent had less than a high school education or GED. I'm going to start with the overall picture of protective orders, and then move into contextual differences. I don't really want to talk about rural urban differences. But some of these I just can't help myself. That education rate did not vary by the rural urban. Forty-eight percent were unemployed at the time of the baseline. However, only 26 percent reported they had not had a job in the past year. And their median income was about $15,000, and that did not vary by area either.

In Kentucky, we don't give protective orders for dating violence. You have to have been married or were married to that violent partner, or to have cohabitated at some point. So it was about 50-50 here — married, cohabitated. And you can also have a child in common. Most of them are cohabitants or married, but about half had children in common with that partner. Just a little bit about the respondents. On average, they were 35 years old. And this is what I really want to hit. I think a lot of you know this is not new in this study, but I want to be very clear here. Seventy-eight percent had prior misdemeanors or felony charges, and of those, there were nine charges on average. Sixty-three percent had prior convictions, and of those with convictions, the average was about seven convictions. Sixteen percent had prior DVOs against other victims in Kentucky because I couldn't get data from other states. And this is from victim reports — 87 percent had ever been in jail; 25 percent, ever in prison. This is one place that I want to point out a rural-urban difference — 35 percent in the urban area had ever been in prison, versus about 15 percent in rural area.

Do protective orders work? I can come up with at least five different ways to measure that. Are there as violations? If there are violations, is the abuse reduced? Can we see a difference? What is the most associated with violations? We talked about that. How do they impact the victim? In particular, fear of future harm and the victims' perceptions of the protective orders effectiveness.

All right, first question, do they stop the violence? What we found is that half of the women experienced a violation in our six month follow-up, and half did not.

Now I want to look at the half that did experience a violation and see if it did reduce violence.

This first bar shows six months before, and this is six months after. You can see there are drastically fewer women. Of those who experienced at least one violation in our six month follow-up — that experience: threats to kill or harm, moderate physical, severe physical rape or threats, or use of the weapon — and of those that experienced any of those tactics, we looked at the number of times, the number of threats before and after, and those are significantly reduced as well.

So what about the abuse — things that might not technically meet a violation or that prosecutors would maybe take on as a violation? Psychological abuse, any economic … what I call symbolic violence — shaking their fist or throwing something or pets, child interference or threats, about to or assault of close others. Again we see a significant drop, and then it's the same pattern. Of those that experienced any of those abuses, tactics, the number of days they experienced abuse were much less from pre to post.

So then we did a couple of multiple regressions to look at what factors are associated with protective order violations. And so we wanted to account for as much of … You go to the literature and some of the studies say children in common was a big difference, his criminal history involvement, prior severity of physical violence, sexual violence and all of that. So we accounted for all of these things of the perpetrator — the area, his age, race, whether or not they had been married to their partner, the number of years in that relationship, whether they had kids in common, and how many days they had spent in the relationship after the protective order — because you might hypothesize that he would have more access, and that would increase violations — physical abuse, severity, sexual abuse and stalking. And what we found both in predicting any violation and the number of violations is that stalking is strongly associated with violations. Just to give you an example — there are two kinds of violation groups, and it was about 50-50. Fifty percent of the women who said there was a violation reported stalking; 50 percent did not. So of those who experienced any violation but no stalking, they reported an average of seven violations, versus 18 for those who reported violations and stalking.

Let's look at fear of future harm. We adapted Dutton's fear scale, and so it looks at fear in a variety of different domains — threats and harassment; fear of physical injury; fear of public humiliation, which is a big one; fear of financial interference; fear of control, continuing to control her life; fear that he would hurt or harass others; and child interference for those who had children under 18. And you can see from before to after, fear levels are greatly reduced. However, let's look at it by violation group. So that first bar — there are no violations. The second bar — there are violations, but no stalking, and then you see those being stalked have significantly higher fear levels, and that makes sense.

Perceived effectiveness. This is no violations and any violations. The vast majority of women felt that a protective order was fairly or extremely effective. But again, when you look at the three groups, you can see that the stalking victims felt that it was less effective. Even though almost 70 percent are still saying that it's fairly or extremely effective.

All right, so that's our five questions. So then what we did was turn to the economic analysis. I worked with the economist and the literature to figure out what costs do we want to include. So the first one was service use — health, mental health, victim services, for example. A big cost associated with intimate partner violence is — whether or not there is a protective order involved — the criminal justice costs — her talking to the police; whether there are charges, convictions; whether he spent any time in jail; and private attorneys and legal aid were incorporated. What we call lost opportunity — anytime she lost from work due to the abuse, anytime she lost civic or family responsibility time and also we looked at volunteer activities that she might have missed — it was very low. We also took into consideration transportation costs and property loss due to the abuse.

And then finally we look at this health-related quality-of-life loss. The reason we were adamant about incorporating that into the cost and all of you who work in the field know why. But methodologically, the way we collected this cost data is by taking a calendar, and we had the participant for each month — six months before and six months after — identify what are the important dates to you? What holidays and birthdays and when do kids start school — anything they could anchor. And then we have her mark her service use due to the abuse because we were trying to partial out preventative health visits and all that kind of thing — trying to keep it specific to partner violence mental-health each month, how many times in each month. And we also did that for a number of tactics, and what you saw on this calendar is … Yes, for some women there was a consistent dark line across for moderate or severe violence, but for most women, that was sporadic. What was black lines all the way across was psychological abuse degradation, humiliation, just this daily grinding her down basically, and so I took those calendars to the economist and said we have to account for quality-of-life. We ran the analysis both with quality-of-life incorporated and not incorporated, but I'm only going to really talk about it with that. And the other reason for looking at that diminished quality of life for the number of days you're stressed, oppressed or anxious because of the abuse is because it is a major, chronic stressor, which we know is related to health and mental health, substance use, lack of sleep, lack of nutrition, and service use.

And then we had to go back and cost out what is the cost of a protective order, and we had done this before in the drug court study. So what we basically did is take that and redo it, and we broke down every single step in the process. In Kentucky, which is very similar to other states, you first go down and file an emergency protective order; it's good for about two weeks. During that time, a hearing is set; usually it's set right at the time of petition so she knows, but then the respondent has to be served with that protective order, and then they come to hearing for the domestic violence order. So we looked at that and the number that had more than one EPO, how many EPOs on average, and we also incorporated amendments so we did the same thing and broke it down. And we worked with the administrative office of the court, the sheriff's office and the court system just to get all the costs right. And they were pretty much the same as what we have found before.

So when the economist got done with the analysis, what we saw, per petitioner, was a difference of about $4,500, and the overall cost of the protective order was about $354. So when he took that information and translated that … So for every dollar spent on the protective order and intervention, there was about $30.75 in avoided costs to society; this is including that quality-of-life indicator.

But then I broke it down by violation group, and you can see that it's a much lower return for those who experienced stalking.

And then what we did is extrapolate these findings to the state population of protective orders that were issued in fiscal year '07. We adjusted for 15 percent for male victims or same-sex victims because their service use and all kinds of things could potentially be different, and when we did that, we found that in a one year, protective orders saved the state about $85 million. Again when you look at it by group and there's stalking involved, there's a much lower return.

So in summary … Protective orders are making a difference in many women's lives. I feel strongly that these data show that they cost very little compared with the costs that victims are bearing and that partner violence is costing society. Again, we just talked about this — they saved the state about $85 million. And we think this is an under estimate actually. We saw that victims appreciate the order, and by the way, only 4 percent dropped the order during that six-month follow-up. Partner violence appears to be part of a pattern of criminal behavior; it's not necessarily an anomaly for many people. And stalking is a huge risk factor for not only ongoing violence but also costs to the justice system, to victims and to society. (27:45)

So what about community context? Before we get into this because I'm going to look at the similarities and then look at the differences between rural and urban. And I talked about this before, but what was really important for us to do was to compare — when you look at rural trends, what I've come to believe, is that you have to have some kind of comparison in order to say this is uniquely different. Also, I think you have to look at multiple kinds of data to really feel clear that there is a difference — sort of data triangulation.

Similarities? There weren't differences by age, income or education; we kind of talked about that. Both groups experienced significant barriers, as noted by the key informants as well as the victims. There is a severe history of violence in both groups. There were no differences in perpetrator criminal justice system involvement history. There were no differences in the effectiveness of the protective order — it didn't matter whether they were in a rural or urban area — half experienced no violations. Stalking came out as a strong predictor, and there were no differences in rural versus urban by perceived effectiveness. Enforcement could have been stronger in both areas, and this is from both the victims as well as the key informants. Although there were some little costs that did differ by rural and urban, the overall costs did not vary or differ.

What are some differences? Rural women wore more entrenched in the relationship; they were more likely to have been married that partner; they were more likely to have children in common with that partner; they were in a relationship longer; and they were more likely to have been unemployed at the baseline interview and to have had no jobs in the past year.

What's interesting to me, too, that kind of gives you a picture of how the process works in a rural area … At the EPO incident more rural women were separated at the time of that incident. More rural women stayed separated from that partner during our six-month follow-up. And of those that did get back together in that relationship, they were in that relationship for fewer days on average than in the urban area.

Again, both groups had severe history of violence but there are two main differences: In the rural area, it looks like the perpetrator just brings people into it — these are child interference threats, threats to harm others or actual threats toward others. And weapons — threats with a weapon and actual use of a weapon — were different in the rural area.

Barriers? I'm going to talk about barriers from the key informants ... and this isn't really surprising if you know the cultural history in the Appalachian rural area — drug- and substance-related crimes and abuse were the number one priority — consistently. Not just from criminal justice system representatives but also from victims services, and that was consistently different from urban areas where partner violence was really high, and I asked it a lot of different ways. There is also some indication from the key informants — and I'm going to show you some secondary data that also confirms these data — that access to protective orders and protective order enforcement is more difficult.

First of all, there is a perception in both groups — victim services and criminal justice — of the women who petitioned for an EPO, on average, fewer would receive the actual DVO than in the urban area. And there were other kinds of barriers that were more negative and blaming attitudes toward victims and about partner violence, sort of more traditional gender roles. Politics plays a much bigger role in the rural area — sort of who you know; the good-ole-boy system came up as a theme. And enforcement seemed to be a lot less asserted.
 
These are some actual data, so this is nonservice rates from the Kentucky State police. What that means is the rate at which EPOs were not served to the respondent. Across the five years in the urban area — I collected five years just to get a more stable picture of this — 22 percent in the urban area were not served compared with about 55 percent in a rural area. And let me say we had four rural counties, and that ranged from 22 percent in one county up to 80 percent in another county — a big county, not a remote, little county.

And charges? What we tried to do to adjust for the population differences is look at the rate of a charge per 1,000 people. We didn't see a big difference for assault in the fourth-degree domestic violence; however, we saw a difference with protective order violation charges, about .49 per 1,000 people in the rural area versus 1.1 in the urban area. And the stalking charges were so low that we didn't do it by 1,000 people, but nine versus 87.

So then I looked at some disposition data from the administrative office of the court and this is the urban area. This is our, the first charge here is assault for domestic violence, and this is about 61 percent with a guilty. Violation of a protective order is this middle one — about 71 percent. And stalking in the second degree, which is a misdemeanor charge in our state, you can see that we're not doing as well; more are dismissed and less are found guilty (about half).

Here's the rural area, you can see by looking at this … these two are the dismissals, so across all three kinds of those charges, much fewer are guilty and 0 percent found guilty for stalking charges in the rural area.

So just victim barriers, we also found some significant differences. Urban women talked about trouble navigating the system, understanding, confusion, not really knowing who to ask questions. Urban women also talked a lot more about the fear and the actual fear of facing the perpetrator in court than rural women. Rural women talked about getting more of the runaround — they actually had to carry their papers to the sheriff's office to have it served, or they had to go find a judge themselves, so they actually had a lot more legwork than urban women and more problems related to politics and judge rudeness. And just as an example, rural women spent about 3 1/2 hours on average to get the EPO verses about an hour and 42 minutes for the urban women.

And another kind of trend, 90 percent of rural women received no contact orders versus 60 percent of urban women. And this comes up as an important point a little bit later — that trend.

What about fear of future harm? And this is at baseline; you can see across every single one of the categories rural women are much more afraid, and they had more sustained fear at follow-up. It is lower, but, it is still higher than in the urban area, but it is lower than obviously from baseline to follow-up.

Let's look at some enforcement numbers. OK, for the EPO incidents, 74 percent — this did not differ by area — reported they talked to the police about the incident. Forty-five percent of urban women perceived or reported there was an arrest for that EPO incident versus 18 percent of the rural women.

Violations, like I said, did not differ; however, the average number of violations differed. There was an average of about nine violations in the urban area and 15 in the rural area. And so, of any violation, 65 percent of women who experienced a violation said they reported that violation. And we did a rough estimate: Urban women reported about an average of 74 percent of violations, while rural women reported about 31 percent.

This is sort of the funnel. Fifty percent experienced a violation; 65 percent reported it. Twenty-seven percent overall were arrested — so 63 percent of the urban and 37 percent of the rural. And then of those that perceived there was an arrest, in the official data, 92 percent, there was actually a charge — a domestic-violence related charge — versus about 43 percent of rural. So overall, 38 percent of urban victims with reported violations believed there was an arrest versus 19 percent of rural victims.

So this next slide shows official data, and it's only of those who said they reported violations. Then I looked at their official data. So what it does is take out whether or not she perceived there was an arrest — so that's the difference between the two slides here. Half in the urban area said there was any kind of domestic violence charge — narrow — assault for violation of protective order or any stalking charge versus 5.6 percent in the rural area.

All right, so enforcement. And then we took the violations and violations plus stalking, and this is of those who said they reported a violation; 50 percent of the violation group showed up with some kind of narrow domestic violence charge in the official core data, and 64 percent of those with stalking had some kind of official domestic violence charge in that six months versus 15 percent in a rural area with violations, but 0 percent of the violations plus stalking in the rural area.

We asked why do you think — if there were any violations in any period that we talked to them about that were not reported — we asked them why did you not report? And this is the table showing for the violations-but-no-stalking group and violations-plus-stalking. And you can see that the majority of women if they didn't report, they said it because they didn't think the criminal justice was going to take it seriously. However, the violations-but-no-stalking group, 43 percent said because it wasn't that serious. But that's not really what the stalking victims say; they're saying there was no point or there was no proof, and there's a lot of bureaucracy.

So in summary … Actually I'm going to skip this because I'm running out of time, and I want to get to some of the … Well, here's one discussion question I still don't know how to talk to the criminal justice system, but if you have any ideas about how to do that, we'll talk about that.

So there was a couple of gaps in victim safety that we identified. One in particular was this — between the EPO in the DVO – it was a mess. It was a mess to tease out. Women didn't realize that first of all it was no contact; they didn't know when he got served; they didn't know what constituted a violation. A lot of them needed to talk about children so was that counted as a violation or not a violation? They didn't know how to report — should I call the police; should I wait until the hearing? So it was just a huge confusion for them. So we came up — Teri and I and the other co-authors — came up with a couple of ways for us to look at that, to better explain the process at the EPO point. Follow-up or more assertive advocacy at that EPO — in between EPO and the DVO — if you're having a horrible time at the EPO, why would you come back for that DVO? If it's not working for you or you don't understand why would you? And of course we have this great thing now — but unfortunately, we didn't have this at the time of the study —  was a protective order notification that will let women know when he gets served if she registered.

Of course, enforcement needs to be strengthened across both areas especially the rural area. And we also asked victims why do you think he didn't violate the protective order? And the vast majority believed that he was afraid of getting in trouble with the law. So the teeth are there, but if he gets away with it one time, then he becomes potentially more empowered. There were three main situations where enforcement was especially difficult: when there's confusion about the primary aggressor, when the offender has fled the scene and in cases of stalking.

And of course, stalking, stalking, stalking ... We need to address it at all levels not just with the criminal justice system but at all levels. These women were very frustrated that she didn't know what to do; she was at a loss. Sometimes they got the same old, same old advice like change your routine. Well in a rural area that's kind of hard to do. Even in an urban area, if you work and you have kids and you're in school — I mean it's hard to do. So they were frustrated at all levels.

And also I want to point out not only was there more violence and abuse, more victim fear and distress and more victim property damage — they don't really show this, but in another analysis I showed that those with violations but no stalking had on average about $15 in property loss during the six-month follow-up versus $600 for the women — this is per petitioner — for women who experience stalking.

And of course, we addressed barriers to navigation and confusion. Gatekeepers was a big one that came up, and I'm not just picking out court clerks — it could've been a police officer; he could've been a judge at the EPO point — the gatekeepers were sometimes really very negative.

And then of course biases that need to be looked at the community level. So, for example, one big one in the rural area is, “Oh if she's dropped that order before, she is not likely to get another order.” And I am thinking, “OK, first of all, if you give 90 percent no-contact orders, she may think she's complying with the law. I don't know if she understands there's two kinds of protective orders you can get. So maybe she's coming back or … Although, we didn't see a lot of dropping in our study, but maybe the women that they see coming back are dropping it because they feel like they are complying with the law, actually. Or, they may be the more severe cases.” So this is a huge bias that needs to be addressed in the rural area. Then things like, “Oh, you know, she's a substance abuser” … and then in the urban area, they had issues like discrimination between mixed race couples and undocumented immigrants. And we saw lots of myths. You know these — I think they're just distractions away from the fact that partner violence is about the systematic and deliberate set of tactics designed to erode someone's personhood, autonomy and to live a meaningful life. So all of these myths or whatever you want to call them — that women do it to gain the upper hand in divorce and all of that — are just distractions from what the real issues are (but I know I'm preaching to the choir).

Barriers have to be connected to the community context, obviously, so how you do that? One way might be to do a walk-through. They're doing this now with substance-abuse treatment, where people are literally walking through the process. Professionals are with you feeling what it is like to be on the other side of this. Or at least to carefully examine at every decision point what are the barriers, the unique barriers with prosecution? What are the unique barriers with police officers? What about the gatekeepers? This is to create systems of feedback and accountability. In Kentucky, we don't have any way for a victim who's been turned away, where a clerk told a woman that the law changed recently, and even though she lived with him for seven years, she is no longer eligible, but that's not true. But there's no way for victims to call and say, “You know this happened to me and is this true? I want to lodge a complaint, but this judge was very rude to me. The judge drug tested me when all I wanted was a protective order.”

And accountability like with what I showed; let us do some data feedback.

And increasing community collaboration, including with researchers. But at that point I'll let Teri take over. [Audience applauds.]

Teri Faragher: (46: 43) I'd like to begin by reassuring Judge McDonald that these studies were not done in his jurisdiction [audience laughter]. He is sitting there sweating; I know he's worried about it, and I just wanted to reassure him … [laughs]. You're welcome.

I'll go ahead and start because you've seen the most important piece here, and there is not a lot of time, and so I really need to get through this quickly, which is why I'm going to stick to some notes here, because if I don't, I know that Bernie wants me to make sure there's plenty of time for discussion and so forth. I echo all of the “thank yous” around this room for everything all of you do, I've come to conferences my entire career to see a lot of people in this room and I can't thank all of you enough because you've been my inspiration for 30 years and the only reason I've been able to continue doing this. And of course particularly, the NIJ for funding this study, OVW, and the National Council for all the support you give to communities like mine. We couldn't do this work without you, and it's just incredibly important, and I hope that's obvious in the remarks in terms of some of the things we're doing.

If you would just go to the first slide, that would be great. So I'm just really going to quickly apply community context. We talked about one of the areas for improvement — stalking, in particular — and I will consider the relationship between research and practice just for a few minutes.

I do want to start by sharing a segment of an article, and I'm not going to read this whole thing, but it was on the front page of our newspaper on September 13th; I'm talking about this September 13th, which is just about a week or two before the study was released on the NIJ Web site. The article was in reference to the murder of a 29-year-old woman named Amanda Ross, who was shot two days earlier, as she left her apartment for work. A 55-year-old former legislator and gubernatorial candidate, Steve Nunn, has been charged with her homicide and both Ms. Ross and Mr. Nunn come from very prominent families in Kentucky. The part that I really wanted you to see … The article begins by talking about how his life went into a downward spiral after the death of his father, the loss of his job and the issuance of the protective order. But the part that I do want to read is that his attorney, or an attorney, Astrida Lemkins, who said she is a friend of Nunn's and was co-counsel for Nunn during a court hearing about the domestic violence order, said the issuance of the order “caused all their problems. It caused Steve Nunn to lose his job, reputation and drove him to slit his wrist. If there does turn out to be a relationship between the death of Amanda Ross and Steve Nunn, it's not because the DVO failed, but rather because the DVO was issued.” Lemkins said, “Ross should have also been held accountable for her role in the domestic violence incident; things are not black-and-white,” she said. “There's a lot of gray in there.”

If you can imagine, the collective blood pressure in our community escalated significantly after these comments were in the paper. And while these comments were very provocative for all of us, I'm sure they were very painful for Ms. Ross' family and friends. They were particularly devastating for victims of domestic violence who were either considering getting a domestic violence order or maybe already had one. This case has received and continues to receive considerable attention for many different reasons, but these remarks in particular did prompt a hailstorm of response from the community, in the form of editorials, op-ed pieces, interviews on talk shows, news interviews, and I'm probably leaving some pieces out — but a whole variety of professionals in the community participated in those interviews and talks; there was a lot of information that went out. It was serendipitous, honestly, that this protective order paper was released right after these comments were in the paper. So these recent data along with the research that Dr. Logan had done for a decade really turned out to inform that discussion, and that was critically important. I've appreciated TK's work for the past 10 years, but I was never so appreciative as I was then.

As you can imagine the information, the data she just presented were not easy to obtain. It's not something I think all of us in this room can take on — it's a huge challenge. And she absolutely gets the lion's share of the credit for everything you saw, but the important piece from my perspective is that it grew out of and was formed by collaboration with community partners; it really was a marriage or cohabitation of research and practice.

In terms of how that helped us inform the discussion, of course, the fact that the protective orders are most effective — that is the most important point that we wanted to get out there — but also just that perceptions are skewed, and TK talked some about that. It's just so important for us, at the local level, to know that our judges, prosecutors, law enforcement officers and advocates know what they're doing makes a difference — that is critically important — right on down to the clerks — everyone. And then also for the victims, of course — it's important that they have good information to make the choices and to know that do protective orders make everyone safer? No, we know that's not true, of course, but we want them to have good information when making the decision as to whether it will work for them or not. And also those statements played into what perpetrators tell their victims — that they are above the law.

I want to talk just a minute about stalking and what this research has meant for us in terms of community practice. I will say that some of what I talk about here certainly … we did get an arrest and enforcement grant from OVW several years ago, and so a lot of our efforts have really grown out of that project.

But in 2005, TK gave a presentation to our local coordinating council (and by the way that is my job — I'm director of our local coordinating council), but TK gave a presentation to my board, this coordinating council, and said that about half of the victims who obtain DVOs are stalked. And I still remember when she said that, sitting there thinking, “OK, well that means 500 to about 750 people in our community — just of the victims who get protective orders — are stalked.” And then I started asking questions about how many charges do we have? How many prosecutions for stalking? And I realized it was literally a handful and that we were really not addressing it at all.

In 2009, I have to say, that was our wake-up call — the alarm is still going off. We know how critical this particular tactic in intimate partner violence is and that we really need to ramp up our efforts to address it. In terms of the kind of things that we've tried to do specifically to address intimate partner stalking, I got some of the things listed there, and I'm not going to go to detail about all of them. But again, we got the grant in 2006, and that was when we started these efforts. Everything takes time to get it up and running, but training was critical, and we put a lot of emphasis on that. It wasn't just individual training of disciplines, but we hosted a statewide conference on stalking — the first in Kentucky — the national Stalking Resource Center came down — they were great, a huge help. We also sponsored roundtable discussions in the community about it, and really, in some ways, that's just the most critical piece, which is getting everyone to start recognizing stalking and being able to tease out stalking behaviors, which leads to the second point — the review of police reports.

We established a procedure where a sergeant in our domestic violence unit who looks at every single police report related to domestic violence. After the fact, if there wasn't a stalking charge on a case in which the behaviors were obvious in that police report, the sergeant would tease it out and slap on that stalking charge later. And then there also was training, back to the patrol, those who maybe didn't identify it to begin with. Periodic reviews with prosecutors and law enforcement of what the trends are and what their arrest numbers are looking like, what's happening with those when they get to the prosecution level, more dedicated efforts to apprehend defenders, and some of this was through that grant, of course. Officers serving as affiants on warrants, which was the sea change for us. Now, some of you may have that in your community, but this meant that when they fled — when the officer got to the house and the offender had fled — because I think they all knew in our community — run— because they are not going to come looking for you. So it meant we had to put more effort into tracking them down, but also it meant that that if they could establish any probable cause, they weren't going to be telling her to go down tomorrow to file the charge, they were going to go and be the affiant on the warrant before they ended their shift.

Frontline advocacy for victims in our police department — we never had that before, so it was trying to offer that option earlier. And then we also established a red-flag committee. We were already having multidisciplinary review of cases that were being prosecuted, but what if they weren't being prosecuted? But any community partner who felt there was an increased risk maybe due to stalking — it could be other issues too; it could be substance abuse on the part of the victim and that would place them at a higher risk. So you could call on this red-flag committee at least to examine the case and see what alternatives they might be able to come up with.

Currently, we have gotten a continuation on that grant and some of our newer efforts are focusing a lot on — just like the grant says — increasing arrests and enforcement. But we have had some success there, but I think we need to look beyond that because not all stalking cases are going to result in an arrest, and not all of them are going to be successfully prosecuted. So what I think we need to do now is to look more broadly across the spectrum of services and be more assertive and creative in what we're coming up with and what else are we doing to help victims be safer when they are being stalked.

We are going to start safety checks. These are victims driven. They can select whether or not, after they've gotten a no contact domestic violence order, whether they want an officer to come by their house and do a safety check once a week, twice a week, for however long they want that to happen or the option of an advocate that is out of our shelter program or domestic violence program. They can choose either one or none.

We are looking at implementing a GPS pilot in Fayette County and TK is very involved with that so luckily we will have good data on that from the start we haven't really launched it yet it's not that — judges can order a GPS now — but of course this will I think intensify that and make it more common practice. We're still working on that though almost as we speak. We still have work to do there, but it has improved tremendously.

And maybe what's more important is the second point that stalking is being recognized by the community partners. We actually have had a number of times upon hearing the facts of the case in court, a judge will say, “Well isn't that stalking? This sounds like stalking behavior to me,” even when there isn't a stalking charge, and this is in protective order hearings that I'm talking about. So it is being taken into account by everyone and particularly by advocates and safety planning.

We all, of course, use multiple processes to evaluate our programs, and we all understand how important it is to do that. This is just a list of some of the things we do, some of the processes we use to evaluate what we're doing, and they are all very useful. I'm not going read through all of those, but generally we do not have the expertise or the time to conduct these kinds of large scale evaluations that you just saw, and I believe that if we are only operating by what we believe works, then we do run the risk of placing victims in greater danger.

I think it is particularly because we all know it is so individualized and what makes one victim safer might place another victim in greater danger, and so the more information we have to help people make those decisions, I think, the better. The relationship between researchers and practitioners are critical because of that.

Of course, I work for local coordinating council, or a CCR, and I personally believe that communities that have those, they do bear a lot of responsibility and hold a community accountable for their interventions. And I also think they provide a logical forum for that communication — bridging that gap between research and practice. TK is on our council; she can communicate with a whole variety of disciplines through that one venue.

I thought a little bit of out what makes this relationship work because I've been in this field for about 30 years, and I've worked with a lot of different researchers over the years who have done projects, but why has this one worked out better than any other relationship, at least that I've ever had, in terms of research and practice. And it really came down to most of those things on this list, and the main one being the first one — the motivation of victim safety, having that trust as the motivation for doing the work. If somebody's doing a research project and they want tenure, I'm sorry I just don't have enough trust ... not that it can't be good, and maybe if it has enough guidance, it might turn out to be great, but for a real relationship for me with research, I have to know that is the motivation and have trust in that motivation.

And also collaboration — I don't have a Ph.D., but I have more experience and practice, so we come to the table with different skills different strengths, but we're at equal footing when we're at the table talking about it, and we can be honest and disagree with each other about things, and we do. We argue quite a bit about stuff, actually. But we hammer it out. And again, it goes back to that motivation, what's your motivation? I do also think that the multiple perspectives — I'm biased toward the advocacy perspective — but I think we have to consider the perspectives of all the disciplines when we are designing these kinds of research projects and thinking about what we will learn from them.

And that last thing on there is always having a standard to measure what kind of difference will it make for victims — is it worth all the effort that's going to go into it? And you know that it's going to make a difference for them.

Of course, there are a lot of challenges in doing this too, and I'm sure many of you have experienced them. Egos, I think, that is the main challenge we all have — with ourselves even — but just in general, it's when you're trying to put something like this together and get people involved — that is big factor.

But crossing over that aisle from academics to practice is a big deal too, and for me to go to that table where I'm certainly less certain of myself and vice versa, you have to be willing to do that.

The fourth bullet down also is a very big one because some people don't want to know what the outcomes are because they are afraid the outcomes might not look so good for them, so you really have to bring along your community partners in that way also.

And then maintaining that balance of influence because neither one should drive the other — they should be informing each other but not driving each other, and that's really critical.

So, in summary, from my experience, this relationship that we've had between research and practice has been very beneficial, and I believe it has greatly benefited victims in our community, and I know it will continue. Honestly, as I was listening to the presentation, I'm still learning from this study; I've been involved with it all along, but I keep finding new things every time I look at it, every time I hear TK talk about it. So thank you so much for your time, and I think now we're going to have to discussion, right? All right, thank you. [Audience applauds.]

Auchter: Thanks to both of you ...

E-mail Moderator: Bernie, I have three questions from the field; can I do those now?

Auchter: OK.

E-mail Moderator: First question comes from New York. Fathers' rights claim that civil protection orders are given out like candy, which you alluded to earlier. In contrast, I've heard women or advocates complain that judges were not taking victims reports of the abuse seriously and frequently denying protection orders. Did they do anything to look at the rates of requests that were refused or rates of conversion from emergency to permanent orders?

Logan: Unfortunately, no. This study again just looked at of those who got the protective order; that is a huge gap, and especially now when we see just how chaotic it was between the EPO and DVO for those who actually went ahead with the DVO — that is a huge gap that we need more information about. But also I think that there is some motivation in some counties to … I don't know what they do with the denied petitions. I think it's hard issue, but it is definitely one that needs to be studied, but it would be like its own devoted study because it would take a lot of time and energy to do.

E-mail Moderator: Thanks. Now, another question from New Hampshire. Did you look at differences between victims who wanted protective orders to ensure no contact and those who wanted continued contact during the PO but felt safer having contact while the order was in effect?

Logan: From my perspective, we've done two studies now … This issue of victims understanding no contact versus no violent contact, it's confusing. And one of the things we saw in the rural area is, although 90 percent were no contact, they would put a note in there, “except for children,” and then women don't always know what it is … So I mean to me, that no contact/no violent contact actually gets really blurry when you look at it over time for victims. So, we did ask a series of questions about victims and their satisfaction with the hearing — whether they could ask questions; the vast majority felt that they had time to ask questions; that the judge asked them, “Do you have questions? Are there needs?” The vast majority said that they felt — again, not all of them, around 70-75 percent ... So we asked a lot of questions — I did not present that here today — around that, about the DVO hearing, but we did not ask specifically, “Did you want a no contact/no violent contact …” because what I find is that women don't really understand it, and even if they do, I feel like the judges kind of, give what they want.

E-mail Moderator: This questioner further asked, “It might be that those not wanting any contact would be more likely to classify contact as stalking, and thus be more likely to report violations. This might inflate the apparent difference in violation rates between stalkers and nonstalkers. What do you think?”

Logan: See in my reporting, and by that I actually mean reporting to the researchers, because they didn't actually officially report at higher rates — the stalkers, stalking victims — and we also asked specific tactics of stalking. The vast majority of these women were being followed, monitored, him showing up at places … So we really looked at his behavior and asked her about his behavior — was it repeated, and didn't frighten her? And that's sort of how we defined it, not whether or not she didn't want contact.

Email Moderator: Question from Maryland, from a public-health person. “You mentioned some legal-system barriers women faced to get protection orders. Did you get a sense from other types of barriers — economic, fear for children, safety, disrupting their routine, etc.?”

Logan: There's a whole lot of inconvenience barriers, and again, our group of women, are women that got protective orders, so they persisted through and put up with the inconvenience and the other kinds of barriers. And a lot of what we talked about when we asked about barriers from the key informants, a lot of them focused on the process and enforcement. Those were the barriers that the key informants focused on and what most of the women actually focused on.

Auchter: Thank you. Questions or comments in the room here?

Questioner:  I was just wondering … Did you look at the remedies that they got into protection order besides no contact, you know like custody or whatever? And did you see any correlation between certain types of remedies that they got and the difference in their outcome?

Logan: We did ask the women what kind of remedies they got, but really, the counties that we deal with, they don't really deal with custody, I mean in general, it's a small percent — they don't really deal with any kind of support. The opportunity is there, but I didn't find a lot of difference between rural and urban, and it was a small percent that did so I kind of focused on whether it was contact or no contact — and then there was such a huge area difference there.

Auchter: Yes.

Questioner:  It's a question about technology and whether when you asked about stalking behaviors, if you also asked about electronic contact and communications?

Logan: We did, but Kentucky is a low-tech state; the stalkers don't need to use technology. I mean, there's a few, there were some e-mail; there were some Facebook slander — maybe one case of that. We didn't have anyone talk about GPS; we had some spyware, but I mean they were very low — more anecdotal — in Kentucky.

Auchter: Are there other questions or comments?

Questioner:  I'm [unintelligible] with the Bureau of Justice and Statistics. I have a quick comment and question about your cost analysis, and it essentially comes down to what do you do with that information, practically? Now in substance abuse treatment, it's pretty clear that treatment is more cost-effective than incarceration, and I think that you can make an argument that that's getting through to policymakers and to criminal justice practitioners, and it may have an effect on their behavior. To play devil's advocate for a moment, I'm not sure if demonstrating that a protection order is cost-effective. Is that going to change the behavior of criminal justice practitioners?

Logan: Well that was one of the questions I started with. The question is how do you talk to the criminal justice system? I came up with the idea of cost — I saw it make a huge difference with drug courts, which are a combination of a justice system intervention. A judge is highly involved in that and there is some treatment. But I mean, I saw a huge transformation with our costs. Why can't it make a difference? Why can we talk about that? Will make a difference? I don't know; I'm asking you — you tell me. How can we talk to the criminal justice system to say let's reduce barriers for women who ask for protective orders? I'm not saying that all women should run out and get one — that has to be her decision. But if she asks, let's reduce it; let's enforce these orders, because you know what? These guys are costing you money. And they are going to continue to cost you money.

Rob Valente: I'm Rob Valente. I'm with the National Network to End Domestic Violence, and I just want to answer and say, I am just so excited to have these data — saying that it does save money because we're facing all kinds of things. The states are cutting budgets, and they are saying it's cheaper to close shelters than it is to worry about protection orders. If there is a specialized docket on domestic violence, they want to close that down and put it in the court of general jurisdiction. And we need these data to explain to them, no, you're actually going to end up spending a lot more money if you do that. It's very important for the state coalitions to be able to share this information out. And I can also tell you that the people who write the check at the federal level, Congress, always ask, “Does it make any difference?” We're spending all of this money to encourage all of these people to have a community coordinated response and get protection orders happening — does it matter? And again, I'm just beside myself. I can't wait to run out of here. I want all of your slides. [Audience laughter.] There is a lot that I want to do with this, so thank you.

Sue Carbon: Hi, I'm Sue Carbon. I'm a family court judge from New Hampshire, and I don't know who that colleague was who was reporting, but I wanted to follow up on an issue that you raised here about the differences between urban and rural. Can you detail a little bit more about why there is such a low rate of women who receive protective orders in rural areas versus those in urban? Can you define a little bit more about what the differences were and why they were so much less likely to receive the orders?

Logan: OK, now this was the key informant … I don't know that; I don't have that data. This was key informant perceptions, so we went through every step of the process … remember how I kept these professionals — judges like you — on the phone for more than an hour, and literally walked them through almost every single process. And so one of the questions we asked, was, “OK, out of 10 women who get an EPO, how many of those do you think are going to end up with a DVO?” And on average, across both the victim services and criminal justice system respondents, I think I said 4 out of 10, versus like, 6 or  7 out of 10 in the urban area. So these were key community professionals' perception of access. I don't have that data. I wish I did. We don't have a system in place that will tell us. That is one of the feedback loops that we need to put in place. I would like a tracking system of every woman who asked for a petition to find out why it's denied. This should just be the state instituted and fed back to the state, which can be used, by county, to look for anomalies and that kind of thing. But we don't have that right now. And right now, like I said, it's almost like they don't have to account for it at all, so I don't know what they do. Do they throw those denied petitions away? I know they're not supposed to, but …

Sue Carbon: It would be really helpful if at some point you could do a study that could look at the content of what the petition is alleging and then look at those which are…

Logan:  I did …

Carbon: You did, OK …

Logan:  I did, between what the EPO and what the women told us in the research …

Carbon: And those which are granted and those which are denied?

Logan:  Oh, no. Yes, well again, accessing the denied the petitions, do I trust that? I'm sorry, judge, I mean maybe you can … [Audience laughter.] maybe you can …

Peter MacDonald: The judge is supposed to give the reason why he — I'm Peter MacDonald, retired judge from Kentucky — why they denied it. And actually if it's denied, it's required by state law to set it for a hearing. I'm sure none of the judges do that though. They still have to come back for a hearing …

Logan:  No …

Peter MacDonald: Oh yeah they do …

Logan:  You don't have to have a hearing…

MacDonald: That's what they think but the statute is pretty clear about that …

Logan:  Right, everybody has a right to a hearing, right. And I understand that. Judge Bowles talked about that too, which was actually a surprise to me — another schooling moment — but what I'm saying is do you agree that potentially there are counties where what happens to those denied petitions?

MacDonald: It depends on where they originated. If they originated in the clerk's office, they should stay there. They should be in the file that is there.

Logan:  Sometimes we can't even find from the clerk's office the actual petitions that were granted …

MacDonald: But it probably just varies from county to county too, about what the practice is — if it's denied — they throw them away; they don't want to save them.

Logan:  Exactly.

MacDonald:  But they should be retained.

Logan: Yes, they should be.

Auchter: We have time for a couple more questions or comments, or reactions, or statements on how this might be utilized before … [Unintelligible.]

Carbon: If I could add just one more thing … I really appreciate this study. This is enormously instructive and informative for us. We have a meeting that follows this to revise and update our Burgundy Book on issuance and enforcement of protection orders, and there is an awful lot from this that were going to take, I think now, and feed into that revised document. This is just incredibly informative. I really appreciate the quality and the time you spent doing this work.

Logan: Oh, thank you so much. I appreciate it. The worst thing for a researcher is for the research to sit on a shelf and never go anywhere, so I always wonder how can I make it more useful. And actually, a lot of the questions that we put on this interview came from the community.

E-mail Moderator: Just so you know … Most of the comments that I got online during this were incredible. Thanks and congratulations, and this is wonderful. Echoing what Sue said, folks out there in the field are just very excited about this, and I think since we've recorded it, one person mentioned they want to listen to it several times and look at your slides and be sure they fully understand this so that they can use it with their own court system. So thank you very much.

Logan: Thank you.

Denise Gommage: I'm sorry if I missed it — I'm sorry, I'm Denise Gommage with the Battered Women's Justice Project — I'm sorry if I missed it … Did you ask about firearms or removal of firearms as part of this process at all, and …

Logan: About … I'm sorry …

Denise Gommage: Removal of firearms as a result of an issued order?

Logan: We did not ask because in most of the counties … [Audience laughter.]

Gommage: It's a religion, I know.

Logan: It's controversial, and even if there was a stamp on it, were they actually removed? And it's just very complicated. Actually, as Teri pointed out in a conversation with Bernie earlier, you know, some victims don't even want his guns removed in Kentucky because if he can't go out hunting with his buddies he's going to be sitting at home and mad. So it's really complicated, so we did not ask about that, but I … it's low. Hopefully, that will change. There is legislation. This high profile case has really spurred a lot of discussion in our state, and they're looking at a lot of changes, and that is definitely one of them.

Auchter: We have time for one more.

Mary Louise Kelley: Hi, I'm Mary Louise Kelley with the Family Violence Program at Health and Human Services, and I think this study is just stunning. I'm really, really impressed and just curious and hope that the information about how you cost it out — services and cost of services …

Logan: It's all in the report …

Mary Louise Kelley: Good, I think we really need to know that …

Logan: It's a 200-page, single-spaced report; basically, it's a book. We tried to be as open with everything that we did. You know, the cost study is not perfect; I'm not going to tell you it is; I mean, it's a start. And to build on that, the economist and I worked really hard to do sensitivity analysis — with and without quality of life — and the 90-percent confidence intervals. We both wrote as much as we could in there — it's a lot of pages, just the cost piece of it, so …

Auchter: Well, thank you all for coming, and thank you …

[Audience applauds.]

Logan: Thank you very much.