The Role of Police Risk Assessments in Judicial Decisions Regarding Domestic

[a]Associate Professor， Deputy Head， Department of Criminology， Ashkelon Academic College， Ashkelon， Israel.

*Corresponding author.

Received 12 June 2013； accepted 22 September 2013

Abstract

Increasing awareness to the issue of domestic violence offenses in Israeli society has led to changes in legislation and enforcement， with an additional degree of severity attributed to domestic violence. These changes have also led the Israel Police to develop an actuarial risk assessment tool to improve the validity of decision-making processes regarding domestic violence. The purpose of this tool was to empower police investigators to assess information from domestic violence complaints， and to derive the best recommended action from that information. Thus， the tool results in uniformity of attitude between various professionals in the law-enforcement system towards domestic violence. To test whether this tool indeed increases uniformity of attitude between various law-enforcement professionals， towards the risk level of the assaulting partner， this study examined all domestic violence offense cases opened in a large city in the south of Israel and analyzed a small sample of protocols from domestic violence investigations that ended with conviction. The study data show that both in requests for remand extension and in penalty judgment decisions， the legal system tends to ignore the risk assessment score provided by the police tool. These data indicate that Israeli legal discourse tends to overlook police risk assessments of domestic violence offenders， which in theory could increase the probability of “false negative” errors in predicting the risk level of an offender. In turn， this may result in additional assaults by violent partners against their victims.

Key words： Domestic Violence； Risk assessment； Court decision

Efrat Shoham （2013）. The Role of Police Risk Assessments in Judicial Decisions Regarding Domestic Violence Offenses in Israel. Canadian Social Science， 9（5）， -0. Available from： http：///index.php/css/article/view/j.css.1923669720130905.2798

DOI： http：///10.3968/j.css.1923669720130905.2798.

INTRODUCTION

Domestic violence offenses have unique characteristics that differentiate them from other violent offenses. These include， among others， the fact that the perpetrators and victims continue， in most cases， to live together under the same roof； the high level of interaction between domestic partners； and the financial， social and personal dependence of the victim on the violent partner. All these factors necessitate that law-enforcement systems attribute greater consideration to a perpetrator’s risk level （Shoham， 2012）.

Over the last two decades， risk assessments have become an integral part of the criminal justice procedure in a number of Western countries （Hanson & Morton-Bourgon， 2009）.

Risk assessments are perceived as essential tools of law enforcement， as they help distinguish those offenders with a higher probability of committing recurring offenses from those whose risk of recidivism is low. The use of risk assessment to both adequate punishment， and to limit and monitor an offender after release from prison， reflects a transition from a doctrine of deterrence to a selective incapacitation doctrine （Sergovitz， 2008）. This doctrine， based on the prevention of future offenses， has attracted considerable criticism mainly because it involves imposing restrictions on the basis of an offender’s affiliation with a particular group or other such characteristics rather than on the actual offense committed.

In Israel， the legal and public discourse regarding risk assessment mainly focuses on sex offenders （Cohen， 2010）. Unlike with sex offender monitoring， Israel does not have legislation necessitating the submission of risk assessments for domestic violence offenders during the various stages of criminal legal proceedings. Never the less， in 2005 the Israel Police decided to implement a computerized inventory tool to assess the risk levels of domestic violence offenders. This tool was designed to quantify the subjective qualitative assessments of police investigators regarding the risk level of domestic violence offenders.

This study aims to examine the extent to which the legal system in Israel adopts the risk assessment provided by the police for domestic violence offenders， both at the beginning of criminal proceedings， i.e. at the first request for remand extension and towards the end of the proceedings， in arguments for sentencing.

1. RISK ASSESSMENTS IN THE CRIMINAL JUSTICE SYSTEM

Domestic violence offenses have unique characteristics which distinguish them from other violent offenses； the fact that the perpetrators and victims continue， in most cases， to live together under the same roof； the high level of interaction between domestic partners； the financial， social and personal dependence of the victim on the violent partner， etc. obligate the law-enforcement system to greatly consider the risk level attributed to domestic violence perpetrators （Smedslund et al， 2007； Campbell， Jones， Dienemann， Kub， Schollenberger， & O'Campo ， 2003； Westbrook， 2007）.

Over the past two decades， risk assessments have become an integral part of how law enforcement and rehabilitation professionals deal with offenders in western countries （for an example of a risk assessment tool （SARA） in cases of domestic violence， see Kropp & Hart， （2000）.

Risk assessments seek to predict the likelihood and characteristics of a particular event， i.e. the frequency， means and expected intensity of that event （Andrews， Bonta & Wormith， 2006； Douglas & Skeem， 2005）. They are seen as essential law enforcement tools， because they help distinguish between those with a high probability of reoffending and those whose risk of reoffending is low （Weinstein， Dayan， Morag， Ziv， Agamy and Mishkin， 2004）.

As a rule， risk assessments refer to an individual’s dangerousness level with respect to a specific potential victim and also to wider society. According to Weiss （2008）， the structured tools contribute to a deeper， more comprehensive understanding of the factors that induce violent criminal behavior； they provide a uniform language that defines “violence” and may therefore help bring more accurate and professional judicial and treatment decisions. The use of risk assessment in the criminal justice system， whether for determining punishment， setting a prison term or to establishing the limitation and supervision levels needed after release from prison reflects a transition from a deterrence doctrine to a selective incapacitating doctrine. This doctrine， based on the prevention of future offenses， has attracted considerable criticism mainly because it involves imposing punishment on the basis of an offender’s affiliation with a particular group or other such characteristics rather than on the actual offense committed （Sapir， 2008； Sergovitz， 2008）.

The increasing frequency of use of risk assessment actuarial tools has also led to different criticism. One type of such critic claims that by treating an individual as an object of scientific study， the assessment focuses on a particular episode of that person’s life and on specific aspects of his personality and actions， while disregarding the wider social contexts and life circumstances in which that person has acted and made decisions （Weiss， 2008）. Most professionals make frequent use of risk assessment tools， but these are often developed overseas and so are not adapted to local populations （Eisenstadt， 2007）.

Ignoring an offender’s total life story presents him as someone who acts indifferently and without morals. Focusing on an offender’s personality may result in unjust accusations and puts responsibility solely on his shoulders， which may lead to a more harsh approach when deciding punishment （Hacourt， 2006； Petersilia， 2008）. The actuarial tools are not sensitive to change， whether of personality or environmental conditions （Eisenstadt 2007； Weiss， 2008）.

Another harsh criticism arising from the use of this actuarial tool is that there has been a shift towards “actuarial justice”， in which the term “dangerousness” replaces “guilty” and distorts perceptions of fair punishment. Actuarial justice does not depend on whether an offense has actually been committed. Instead， this school of thought argues that， just as it is justified to prevent an offender from committing any future felony， it is also justified to prevent a person who has yet to break the law from performing a first offense especially when considering the severity and harm caused by that offense. Serogovitz （2008） claims that “the fear of crime” in countries such as the United States of America has resulted in a large-scale imprisonment of civilians and has replaced previous policies which called for minimum incarceration as part of a danger-management agenda.

In Israel， judicial and public discourse mainly focuses on risk assessments of sex offenders. The Public Protection from Sex Offenders Act 2006 （renamed during 2011 to the Public Protection from Sex Offenses Act） was designed to protect the public from recurring sex offenses by implementing risk assessments at various stages of the judicial procedure as well as through monitoring and supervision programs. This Act allows various elements within the criminal justice system （such as courts， parole boards， psychiatric committees） to define conditions and periods for monitoring and supervision of sex offenders based on a risk assessment （Shoham， 2008）. Unlike sex offenders， as mentioned earlier， Israel does not have legislation that stipulates the submission of risk assessments for domestic violence offenders during various stages of criminal proceedings.

2. POLICE HANDLING DOMESTIC VIOLENCE OFFENSES IN ISRAEL

Until the 1990s the Israeli legislature did not distinguish between violence towards a stranger and domestic violence； the legal penalty for violence was the same for all offenses. To this day， the Domestic Violence Prevention Act does not define “domestic violence” but refers to existing sections in the Penal Code （1977） which address common assault， assault causing actual bodily harm， aggravated assault of a family member， rape， threats and damage to property （Shoham & Abulafiya， 2010）.

However， Public and parliamentary activity， led mainly by women’s organizations in Israel， led to changes in legislation that attribute additional severity to domestic violence. These amendments were made to create a perceptional transition from prioritizing the protection of the family unity to a more severe punishing of the offender and better protection of the victim.

As a result of these legislation changes， in late 1998 the Israel Police established a unit of specially-trained investigators， domestic violence） DV） Investigators， who were equipped to deal with these unique offenses. The training given to these investigators was designed to reduce prejudice among police officers towards such incidents in general and towards the victims in particular （Shoham， 2012）.

Never the less， according to the 2011 Police Statistical Yearbook， a significant number of domestic violence cases opened by the police， do not result in indictments. The most prominent reason that these cases are closed is lack of evidence， and the second most common reason is lack of public interest. The total number of cases closed for various reasons is approximately 40%. Only four percent of all domestic violence cases in a particular year eventually reach court （Almog-Lotan， 2011）. Bar-Eli， Bar-Mocha and Frenkel （2004）， shows that in cases of suspected domestic violence in Israel， police officers often have to choose between alternatives， which requires some level of prediction. They claim that the police tend to choose a strategy that will lead to higher level of false-positives errors， while the courts tend to adopt a more conservative strategy which theoretically increases the probability of false-negative errors in risk prediction.

3. THE POLICE RISK ASSESSMENT TOOL FOR DOMESTIC VIOLENCE OFFENSES

As part of the change in attitude towards domestic violence within the law-enforcement system， the Israel Police have developed a risk assessment tool for domestic violence offenders which is designed to help police investigators reach decisions in the field. The general approach behind the risk assessment tool was based on three principles （ibid.）：

The tool must allow information intake from various input sources on issues that can help form an indication of the poential risk level of domestic violence.

The tool must allow quantitative integration of the material so that a risk index can be calculated.

The tool must be decision-supportive so that a structured recommendation can be based on it.

Two clinical psychologists， a social psychology researcher and three senior officers from the Israel Police investigation department， with vast experience in domestic violence cases， participated in the process of developing and validating this tool （Weinstein， Dayan， Morag， Ziv， Agamy & Mishkin， 2004）. The final tool is comprised of measurable and structured risk assessment parameters， including：

Biographical characteristics.

Specific violence characteristics the characteristics of the offenses such as assault， threat， firearm possession， suspect characteristics and the relationship between the suspect and the victim.

A history of past risk assessments.

Each item is given a relative weight and the general index is calculated according to the sum of weighted items （Karon， 2007）. The following table shows the scoring range and allocated risk level division， including recommended action decided according to risk level：

The purpose of this study， as mentioned above， is to review whether the implementation of the police risk assessment tool has led to a convergence in decisions between the various elements of criminal proceedings for domestic violence offenses. To do so， the study focuses on two main questions：

Do judges refer to the risk assessment submitted by the police when making initial remand extension decisions for domestic violence suspects？

Do judges refer to the police risk assessment when determining the penalty judgment in domestic violence offenses？

This study hypothesized that there would be a tendency toward incongruence in the various strategies taken by the courts compared to those taken by the police with regard to risk assessment scores produced by the Israel Police tool for domestic violence offenses.

4. METHOD

To examine the role plays by police risk assessment in judicial decision-making， all cases in which a domestic violence complaint had been made and an indictment was filed， in a large police station in the south of Israel between 2006-2008 were collected. Due to the large number of domestic violence cases opened each year， the sample included all the files opened at the beginning， middle and end of the year， i.e. for the three months of January， May and August for each year.

To do so， the cases relevant for this study were separated from the total cases. Out of 4，061 complaints regarding domestic violence incidents registered in the police station in 2006-2008， only those cases concerning violence towards a female partner were selected a total of 2，076 criminal files. Of these， only cases where an indictment was finally filed during the study months were added to our sample. The following details the final distribution of relevant cases during the selected period：

2006： 61 cases resulted in an indictment， which are equivalent to 31.93% of all cases opened during January， May and August.

2007： 65 cases resulted in an indictment， which are equivalent to 28.13% of all cases opened during January， May and August.

2008： 66 cases resulted in an indictment， which are equivalent to 27.89% of all cases opened during January， May and August.

To obtain the data for this study， permission was obtained from the Israel Police to review the cases stored in the computerized system and take out cases from the archive.

All of the parameters detailed in the police investigation cases were then divided into two general categories： those related to police handling of the case and those related to the judicial procedure.

Parameters related to police handling of the case： the victim’s gender， is there an available risk evaluation （yes/no）， the risk level， complaint submission date and risk assessment submission date， the gender of the serving judge during the remand extension， the offense severity （from property damage to rape）， the number of remand days （the police request for a number of remand days compared with the remand period decided on by the court）.

Parameters related to the judicial procedure： court ruling （guilty or innocent）， plea bargains （exists/does not exist）， offense severity （from property damage to rape）， the judgment （from obligation to avoid repeating the offense to prison sentencing）.

5. FINDINGS

5.1 The Role of Police Risk Assessment During Remand Extension Request Stage

The first research question centered on the relation between the court’s response to the police request for first remand extension of domestic violence suspects and the risk assessment score provided by the police for the case in question. When handling remand extension requests， the court may choose one out of four alternative options： remand， release with bail， remand until end of criminal proceedings and mentally unfit for trial.

Table 1 shows the average risk assessment score for each of the four judicial decisions （remand， release with bail， remand until end of proceedings and mentally unfit for trial）， during the first remand extension request stage.

Table 1

In each group the average risk assessment score was greater than 100. According to the domestic violence offender risk assessment tool developed by the Israel Police， when the risk score is greater than 100 the suspect should be brought before a judge to request remand and consider remand until the end of proceedings （release with bail， for example， is an option that should be requested when the risk score ranges between 0-40）. Although an analysis of variants test （ANOVA） showed a significant difference between the average risk scores of three types of judicial decisions （f（2）=12.36， p

In light of these findings the study continued to examine whether there is a significant difference in the total remand extension period the police requested and the total remand extension period finally decided by the judge. A t-Test for Independent Samples found a significant difference between the average number of remand days decided by the judge （2.33 days） compared with the average number of remand days requested by the police （3.66 days） （T（95）=12/17， p>0.00）.

5.2 The Role of Risk Assessment During Penalty Judgment

Out of the total number of penalty judgments decided on by judges in the cases reviewed in this study， in 52% the defendant was handed down a suspended sentence， in 22% cases he was sentenced to prison terms of up to six months （which were converted to community service hours）， 13% of cases were sentenced to prison terms of more than six months and 13% of defendants were fined.

The second research question asked whether judges refer to the score of the police risk assessment tool in their judgments.

To perform quantitative analysis of the cases， 15 cases （five of each year between 2006-2008） were selected. The sampled cases were concluded and， in addition， included a risk assessment carried out using the police tool.

Table 2 details the judge’s reference to the police risk assessment attached to the case as well as reference to the victim’s testimony regarding the violent offense for each of the cases we selected.

Table 2

None of the reviewed cases shows that the judge made reference to the risk assessment score when deciding on the penalty judgment.

This is despite the fact that， aside from two cases， all other cases including injury/rape accusations had a medium or high risk score. The sentences handed down in the reviewed cases also do not reflect the police risk score as produced by the risk assessment tool.

5.3 The Impact of the Victim’s Voice During Penalty Judgment Decision

Initially， this study did not intend to focus on the impact of the victim’s voice as a consideration in the judicial decision regarding the penalty judgment. However， when we found that there had been no mention of the police risk assessment in the total of cases reviewed in this section， we decided to conduct a qualitative analysis of whether judges chose to refer to the victim’s voice regarding the violent offense in question.

To do so， protocols from 30 trials were analyzed to examine whether judges referred to the voice of the victim， i.e. the victim’s fear of the defendant in the future， their desire to restrain the convicted partner or as found in many of the sampled cases to keep the defendant within the family.

The cases sampled from 2006 had risk assessment scores ranging between low and high. All cases revolved assault offenses. A reference to the victim’s testimony was found in one case： “…reviewing the victim’s testimony shows that the woman fears the defendant…”. This case had a high risk assessment score （155） and the defendant was sentenced to a community service penalty up to six months. In the rest of the cases， the judges did not refer to the risk assessment or to what the victims had to say， and the penalty was fine.

Ten cases from 2007 were also sampled， all regarding an assault offense （one included a violation of a judge’s order）， each with a different risk assessment. Three cases referred to the victim but not the issue of dangerousness expected for the victim. The protocol of the case with the low risk assessment score （21） shows that while the judge did refer to the victim’s voice in this case： “…I noticed that the plaintiff was very agitated during her testimony…” there was still no reference to the risk assessment score. In the case with a medium risk assessment score （62）， the judge referred to the victim’s probation review but the sentence imposed was a fine， even though the judge quoted from the victim’s testimony showing the need to restrain the defendant from his partner.

The review of court protocols undertaken in these cases showed that， out of ten cases reviewed for 2008， the judges referred to the victim’s voice only in two cases； in one， the judge referred to the victim’s testimony： “…from the victim’s probation review it seems， and after hearing the victim’s testimony…”. In the other case the judge referred to the victim’s voice when testifying at the police station： “…the plaintiff detailed what she has been through in the apartment she rented with the defendant…”

DISCUSSION

Towards the end of the 20th century， wide-ranging public activity aimed to increase awareness towards the issue of domestic violence in Israeli society has resulted in changes in legislation designed to create a formal response that focused on punishing offenders and protecting the victims rather then protecting the family unity （Shoham & Abulafiya， 2010）.

These changes have led the Israel Police to also develop an actuarial risk assessment tool to assist police officers in reaching more structured and valid decision regarding domestic violence， and strengthening uniformity between various professionals in the law-enforcement system towards these offences.

Never the less， our data shows that the judicial system tends to dismiss police risk assessment scores in domestic violence cases. These findings support those of Bar-Eli， Bar-Mocha and Frenkel （2004）， and point to the fact that while the police system aims to protect the public， thereby creating a higher probability of false-positive in predicting dangerousness， the courts focus on personal impressions from the circumstances of the event and from the defendants themselves， increasing the probability of false-negative when performing non formal risk assessment （for more details on the parameters weighed in judicial decisions for domestic violence offenses see： Shoham & Abulafiya， 2010； Agmon-Gonen & First， 2007）.

Although the average risk level was significantly lower for suspects released on bail than for suspects who remained in custody or who were remanded until the end of criminal proceedings， according to their risk-assessment tool score， even this group was supposed to be remanded in custody. Regardless of this recommendation， the court’s judgment was to release them all the same. Sandberg （2000） explains that as a rule， offenses that indicate the dangerousness of the suspect include violence or threats of violence and danger to life or bodily harm. However， the severity of the offense is not a reason for detention and so， even in severe offenses police officers must convince the court that the suspect is indeed dangerous. Questions regarding dangerousness focused on whether there is factual evidence from which one can infer the existence or absence of such a rationale. Gilboa （2008） argues that the severity of an offense should be differentiated from dangerousness. In his opinion， one should distinguish between dangerousness based on what he refers to as “objective facts”， such as past convictions for violent offenses， multiple offenses， firearm possession， etc.， and dangerousness based on the fears and subjective concerns of the victim， the witness or the police officer.

Zakay and Fleisig （2010）， who reviewed literature concerning the main characteristics of judges’ decisions and the degree to which they are free from error and bias and conform to normative criteria of rational decisions， conclude that judicial decisions do not always align with the requirements for rational decisions. According to them， one of the causes of this is heuristic reasoning which allows for thought-processes that are fast， automatic and have a lack of awareness. They claim that such reasoning leads， in a number of cases， to stereotypical reasoning and cognitive bias.

This study’s findings show that at the stage of the first request for remand extension， judicial decisions cannot be predicted based on the police risk assessment. This indicates that in domestic violence offenses， the judicial system wishes to distinguish， even in such early stages， between the severity of the offense and the dangerousness attributed to the suspect by the police. It seems that the domestic violence offense risk assessment calculated by the police tool begins and ends with the police， according to a senior psychologist working for the Israel Prison Services （in an interview conducted in June 2011）. “…if the judges were to refer to the police tool’s risk assessment score they would all be in custody and the system would collapse”.

Indeed， as seen in other Western countries， “actuarial justice” has led to the undermining of social services systems （Sergovitz， 2008）. The psychologist added that， in his experience， if the judge believes that the suspect is indeed dangerous he will ask for a risk assessment to be carried out by a professional and will not rely on the police tool’s risk assessment. This conception is also supported by other informal interviews with several investigation officers from the police station used in this study （30.10.2011）. The officers interviewed claimed that they have yet come across a judge who made reference to the risk assessment in requests for remand extension.

In addition， the officers claimed that positive answers to critical questions in the assessment tool questionnaire automatically increase the risk level to high， and so it does not permit differentiation between different levels of dangerousness. It is possible， therefore， that the lack of correlation between the dangerousness attribution produced by the police tool and decisions regarding suspects’ remand are not just an expression of two different strategies but also the result of an actuarial tool which structurally attributes a high risk level to suspects with a differential risk level （for an example of a description of the use of different tests in the American criminal justice system， see Harcout， 2007）.

Although this paper did not directly address the judicial system’s reference to the victim’s voice （for more about this issue see Dancig-Rosenberg & Pugach， 2010； Shoham & Regev， 2008） and despite the small number of protocols analyzed （only 10% of the cases in the study sample）， the dismissal of the victim’s wishes correlates with those of Shoham and Abulafiya （2010）， who investigated judicial decisions in Israel Magistrate’s courts for domestic violence offenses. The researchers in that study concluded that， despite changes in legislation and enforcement， domestic violence offenses are still not considered exceptionally severe or dangerous. This conclusion is based on the large number of cases closed by the State Attorney or the court （approximately 40% of indictments are canceled or suspended）， the long intervals between the time the offense occurs and the end of the trial， and the lenient penalty judgments （actual prison sentences are handed down in less than 10% of cases） （to read more about the high rates of closure of domestic violence cases see also Almog-Lotan， 2011； Shoham， 2012）.

Classic criminal law puts the suspect at the center of the criminal proceeding， thus contributing even further to the exclusion and silencing of the female victim. This leads to decisions that， in many cases， contradict the woman’s wants and needs. Hadad （2012） shows that it is rather when a judge mentions the victim’s desire for domestic peace that harsher punishments are given （a prison term） compared to similar penalty judgments where there is no reference to the victim’s desire for peace. This problem is exacerbated in the context of Israeli society’s multicultural background. Criminal law’s definition of domestic violence offenses mainly centers on assault offenses and therefore cannot encompass multifaceted cultural complexities of the domestic abuse phenomenon （to read more about the cultural facet of handling with domestic violence offenses， see Shoham， 2012）.

Another interesting finding arising from this study is the significant difference in the length of remand extensions for domestic violence suspects as sentenced by female versus male judges. These findings match those of Bogoch and Don-Yehiya （1999）， showing that it is actually female Magistrates’ court judges who hand down lighter sentences for domestic violence offenses than do male judges. One possible explanation may be that women judges are internalizing male value systems and feel a need to prove themselves to the male system and to men around them. Another explanation for this phenomenon is that women feel a subconscious need to detach themselves from female victims and so they prefer to blame the women and assume that something in their behavior was improper.

The Israeli judge Zlotchover （2004） argues that in the emotional heat of addressing the difficult issue of domestic violence one forgets the basic rights of the suspect， detainee or defendant. The judge rejects decisions on the grounds of severity which has more to do， in his opinion， with public interest and less with the moral decision that the judge must make concerning the benefit of prosecuting the partner charged with a violent offense. The judge is dubious about attempts to use the court to change public opinion and turn domestic violence into a social problem.

The judicial system’s dismissal on the one hand and the high risk score given to a significant number of domestic violence suspects on the other raises the probability of “false alarms” while simultaneously “missing the target”， thus harming the capacity of the system to efficiently obtain a reduction in domestic violence offenses and to provide adequate response and protection according to the needs of victims of such offenses.

Either way， at this point it is impossible to tell whether the court’s dismissal of police risk assessments stems from its general perception and preference to deal with evidentiary and positive circumstances rather than in educated assumptions of the risk assessment， or whether it is the judicial system’s lack of trust in the actuarial tool developed by the Israel Police.

Background criticism directed at the police that it considers every suspect as guilty， may indeed lead to the fact that the judicial system’s dismissal of police risk assessments clarifies the need for the standardization of the tools used by the criminal judicial system in the various stages of investigation.

This study was based on police data from a single large city in the south of Israel， and therefore this finding should be considered with caution. The judicial system’s reference to the police tool in judicial decisions in other districts should be further investigated. In addition， the police must continue to examine the reliability and validity of the tool it developed to see whether its capacities can be improved to distinguish between the various risk levels of domestic violence offenders.

REFERENCES

Agmon-Gonen M.， & First， A. （2007）. Is the car better？！ Punishment in domestic violence offenses. In D. Barak-Erez， S. Yanisky-Ravid， Y. Bitton & D. Pugach （Eds.）， Reviewing Law， Gender and Feminism （545-582）. Tel Aviv： The Uno Academic Campus/ Nevo Publishing Ltd.（Hebrew）

Almagor-Lotan， O. （2011）. Violence against women Data summary for 2011. Jerusalem， the Knesset Information and Research Center. （Hebrew）

Bar-Eli M.， Bar-Mocha， M.， & Frenkel， A. （2004）. The problem with home restriction in cases of suspected domestic violence： risk-taking perspective. Police and Society， 8， 37-53. （Hebrew）.

Andreus， A. D.， Bonta， J.， & Womith， S. （2006）.The recent past and near future of risk assessment. Crime & Delinquency， 52， 7-27.

Archer， J.， & Browne， K. （1989）. Concepts and approaches to the study of aggression. In J. Archer & K. Browen （Eds.）， Aggression： Naturalistic Approaches （pp.3-24）. London and New York.

Buzawa， E. S.， & Buzawa， C. G. （1993）. The impact of arrest on domestic violence. American Behavioral Scientist， 36（5）， 558-574.

Campbell， A. （1993）. Men， women and aggression. London： Basic Books.

Campbell， J.， Jones， A.， Dienemann， J.， Kub J.， Schollenberger， J.， & O’Campo， P. （2003）， Intimate partner violence and physical health consequences. Archive of Internal Medicine， 162， 1157-1163.

Campbell， J. C. （2003）. Assessing risk factors for intimate partner homicide. NIJ Jornal， 250， 14-19.

Cohen， D. （2010） Risk assessment as part of the handling ofsex offenders. In M. Shchory， & S. Ben-David （Eds）， handling of sexual offenders in Israel Challenges in recognition， characterization and therapeutic response. Jerusalem： Carmel and Keshet Press. （Hebrew）

Dancig-Rosenberg， H.， & Pugach， D. （2010）. When love hurts： the dilemma of considering the request of women living with violence to reduce the punishment of their aggressor. Law Studies， 26（b）， 589-652.

DeKeseredy， W. S.， & Schwartz， M. D. （2011）. Theoretical and definitional issues in violence against women. In C.M. Renzetti， J. L. Edleson， & R.Kk. Bergen （Eds）. Sourcebook on violence against women （3-20）. Thousand Oaks， CA： Sage.

Douglas， K. S.， & Skeem， J. I. （2005）. Violence risk assessment getting specific about being dynamic. Psychology， Public policy and Low， 11（3）， 347-383.

Eisenstadt， M. （2007）. Is the term “Dangerousness” dangerous？ Social Safety， 75， 65-88.

Gal， N. （2003）. Violence towards women： normative or deviant？ Tel Aviv： Hadekel Press Ltd.. （Hebrew）

Gilboa， M. （2008）. X files or well aimed arrows： comments regarding the police representative investigation during arbitrary arrest procedure. The Defense Attorney， 134， 4-8. （Hebrew）

Goldstein， S. （2008）. Reinforcing the position and role of the victim in both civic and criminal proceedings. In I. Kim， Y. Bar Zohar， & L. Eden， （Eds.）， The Victims of Law Enforcement， Sex and Society （pp.170-176）. Tel Aviv： Massada. （Hebrew）

Hadad， L. （2012）. The Israeli court’s attitude towards violence against women： a quantitative study of Magistrates’ court penalty judgments. Seminary project， Ashkelon： the Criminology Department at the Ashkelon Academic College. （Hebrew）

Hanson， R. Karl， & Morton-Bourgon （2009）.The accuracy of recidivism risk assessments for sexual offenders： A meta-analysis of 118 prediction studies. Psychological Assessment， 21（1）， 1-21.

Harcourt， B. E. （2006）. From the asylum to the prison： Rethinking the incarceration revolution. Texas Law Review， 84， 1751-1786.

Harcourt， B. E. （2007）. Against prediction： Profiling， policing， and punishing in an actuarial age. The University of Chicago Press.

Karon A. （2007）. Tools for risk evaluation of violent behavior towards another and suicidal behavior literature review. Retrieved April 10 2011， from http：//.il.

Kropp， P. R.， Hart， S. D.， Webster， C. D.， & Eaves， D. （1995）. Manual for spousal assault risk assessment guide （2nd ed.）. British Columbia： The British Columbia Institute on Family Violence.

Lernau， H. （2008） Recipe for inefficient legislation： The case of the Detention Act. Law Studies， 24（a）. Retrieved May 8 2011， from the Public Attorney’s website on http：//www.justice.gov.il

Petersilia， J. （2003）. When prisoners come home： Parole and prisoner reentry. Oxford University Press.

One in 100： Behind Bars in America. （2008）. Pew Center on the States.

Sandberg， H. （2000）. Offense severity in arrest rationale Is it really the end？ Law， 31， 323.

Sapir， Y. （2008）. Against prevention？ A response to Harcourt’s against prediction on actuarial and clinical predictions and the faults of incapacitation. Law & Social Inquiry， 33， 253-264. （Hebrew）

Sergovitz， M. （2008）. Risk assessment： Theoretical background and criticism. The Defense Attorney， March， 132-138. .（Hebrew）

Shoham， E. （2012）. To glance behind the walls： Domestic violence in closed communities. Be’er Sheva： Ben-Gurion University in the Negev Publishing. （Hebrew）

Shoham， E.， & Abulafiya， Y. （2010）. Judicial decisions in domestic violence offenses at Magistrates’ courts. Trends， 47（a）， 103-129. （Hebrew）

Shoham， E. （2010）. The dark side of the sun. Be’er Sheva： Ben-Gurion University Press.

Shoham， E.， & Regev， Y. （2008）. The victim’s impact statement in Israeli criminal proceedings. In E. Shoham （Ed.）， Supervision of sex offenders in Israel Treatment or therapy？ Tel Aviv， Perlstein-Ginosar Law Books Publishing. .（Hebrew）

Smedslund， G.， Dalsb?， T. K.， Steiro， A. K.， Winsvold， A.， & Clench-Aas， J. （2007）. Cognitive behavioral therapy for men who physically abuse their female partner. Campbell Collaboration Systematic Review. Retrieved September 20， 2007， from http：///doc-pdf/070716_CBT_violence_final.pdf

Tayagi， S. （1998， June）. Risk assessment of male batterers. Education Wife Assault Newsletter， 9

The Israeli Women’s Network. Retrieved June 1 2011. http：//.il

Walker， L. E. A. （1993）.The battered women syndrome is a psychological consequence of abuse. In R. J. Galles， & D. R. Loseke （Eds.）， Current controversies on family violence （pp.133-153）. London： Sage Publication.

Weinstein， A.， Dayan， K.， Morag， M.， Ziv N.， Agamy Z.， & Mishkin， D. （2004）. Domestic violence： A computerized tool for risk assessment. Jerusalem： Israeli police. （Hebrew）

Weiss， Y. （2008）. Structured tools for risk assessment of criminals and sex offenders. In E. Shoham （Ed.）， Supervision of sex offenders in Israel Treatment or therapy？ （pp. 47-58）. Ben Gurion University Press. （Hebrew）

Westbrook， L. （2007）. Digital information support for domestic violence victims. Journal of the American Society for Information Science and Technology， 58（3）， 420-432.

Yassour-Borochowitz， D.， & Lavi-Efrat， Y. （2007） Orders of protections of buttered women in Israel. Society and Wellfare， 27， 237-254. （Hebrew）

Zakay， D.， Fleisig， D. （2010）. Heuristic reasoning and its influence of judicial decisions. Law and Business， 12， 91-116.

Zlotchover， N. （2004）. Legislation trends in the field of domestic violence. Police and Society， 8， 5-37. （Hebrew）