FREE ELECTRONIC LIBRARY - Dissertations, online materials

Pages:     | 1 |   ...   | 5 | 6 || 8 | 9 |   ...   | 11 |

«Evaluation of Three Court-Mandated Family Violence Interventions: FVEP, EXPLORE, and EVOLVE Stephen M. Cox, Ph.D. Professor Pierre M. Rivolta, Ph.D. ...»

-- [ Page 7 ] --

Court-Mandated Family Violence Interventions Central Connecticut State University The findings were similar when looking at family violence arrests (Table 21). The EXPLORE treatment group had a much lower one year arrest rate (16%) for family violence offenses than the EXPLORE comparison group (27%). These differences were statistically significant.

–  –  –

EXPLORE Effect Size Calculations The above analyses found that EXPLORE participants were more successful than eligible offenders who did not attend it. We next present the effects sizes to provide context as to the magnitude of these program effects (Table 22).

–  –  –

The dcox effect sizes for both any arrests and family violence arrests demonstrate that EXPLORE produces moderate effects (-.54 and -.40). Again, the more negative the effect sizes the more impact EXPLORE had in reducing offenders’ recidivism. Similar to the FVEP effect sizes, the 95% confidence intervals provides evidence these effects are statistically significant and can be generalized to all EXPLORE participants. The odds ratios provide context for these effects, in that, offenders who did not attend EXPLORE were more than twice as likely to be arrested as offenders who completed EXPLORE. This likelihood was slightly lower for family violence arrests. Eligible offenders who did not attend EXPLORE were 1.94 times more likely to be arrested for a family violence offense than EXPLORE participants.

–  –  –

EVOLVE is a 26-week 52 session (2-hour sessions, twice a week) post-conviction and post-plea program. It is an intensive cognitive behavioral intervention designed for high-risk family violence offenders (males only). It centers on victims and children, behavior change, interrelation and communication skill building, and responsible parenting/fatherhood.

Matching Process for EVOLVE

Similar to assessing outcomes for the FVEP and EXPLORE, the first step in this process was to create a comparison group of similar offenders to those who participated in EVOLVE. In creating an EVOLVE comparison group we first identified offenders eligible for EVOLVE but not referred. Similar to EXPLORE, EVOLVE is a post-plea program so we first constructed an initial list of eligible EVOLVE offenders comprised of those individuals convicted of a family violence offense in 2010 and sentenced to probation. This list consisted of 1,609 offenders (i.e., no-program comparison group). The referral and selection process for EVOLVE is similar to EXPLORE and eligible offenders may not be referred to EVOLVE due to the same types of pretrial activities that affect EXPLORE referrals. For instance, eligible defendants may accept a plea agreement that does not include a court-ordered probation condition requiring EVOLVE participation; probationers may be ordered to attend EVOLVE but are allowed to substitute nonJudicial programming such as private counseling or therapy; and, probationers may be ordered to attend EVOLVE but have their court-ordered conditions changed following disposition.

Unfortunately, data are not available that would allow us to know the exact reasons why eligible probationers do not attend EVOLVE16.

The EVOLVE participation’ group (i.e., EVOLVE treatment group) and the no-program comparison group were merged to create one large dataset, hereafter referred to as the EVOLVE merged sample. This merged sample consisted of a total of 1,794 cases: 185 EVOLVE participant cases and 1,609 no-program comparison cases. Data in each group were then checked for missing values and 167 cases were removed for missing court or DVSI-R information. After deleting these cases, the data set was reduced to 185 EVOLVE participants and 1,442 noprogram comparison cases (total N = 1,627).

Following the sampling procedure detailed above, propensity score matching (PSM) was employed to minimize selection bias and ensure the subjects in the comparison group were similar to treated subjects on 14 covariates (i.e., age at arrest, racial group membership, court17, DVSI-R total score, DVSI-R risk level, DVSI-R risk to victim, DVSI-R dual arrest, number of prior arrests, number of prior family violence arrests, number of prior jail sentences, number of During and prior to 2010 probation officers rarely added probation conditions for EVOLVE attendance. Since 2010, CSSD policy has been revised to encourage probation officers to refer eligible probationers to EVOLVE without a court-ordered condition.

Note: Recall that EVOLVE is only available in the following four court locations: Bridgeport, New Haven, New London, and Waterbury.

–  –  –

prior probation sentences, number of prior family violence jail sentences, and number of prior family violence probation sentences).18 The process began by aggregating the groups using a custom SPSS plug-in (see Thoemmes, 2012). The program used logistic regression as an estimation algorithm to calculate the propensity score for each subject in the dataset using the 14 covariates specified above. And then, using the nearest neighbor matching algorithm, the program matched one subject in the treatment group (i.e., EVOLVE participants) to one subject in the comparison pool (i.e., noprogram) where the propensity score most closely matched the treatment subject’s propensity score (i.e., 1:1 nearest neighbor matching), without replacement. The program repeated the process until each treatment subject was matched to a comparison subject.

During the propensity score matching procedure, cases in the no-program comparison group that were not matched to EVOLVE participants (n=1,257) were removed from the data set.

The propensity scores matching procedure resulted in 185 no-program comparison subjects matched to the 185 EVOLVE participants. Following the matching procedure, the balance of all observed covariates as well as interaction among all covariates were examined. No covariates exhibited a large imbalance (|d|.25). The overall balance test was also not significant (chisquare = 16.402, df = 28, p=0.96), and the relative multivariate imbalance measure L1 was larger in the unmatched sample (.998) than in the matched sample (.995). These measures indicate the matching procedure successfully improved balance between groups. In addition, diagnostic plots were produced and show that covariate balance was greatly improved in the matched sample. A selection of plots is presented hereafter.

Figure 7 below shows the actual propensity score distributions of both groups before and after matching overlaid with a kernel density estimate.

Note: the EVOLVE program is designed for male batterers only (as is EXPLORE). As such, there was no need to match on gender since all program participants were male.

Court-Mandated Family Violence Interventions Central Connecticut State University Figure 7. Distribution of Propensity Scores for the EVOLVE Study Groups Figure 8 below shows a line plot of standardized differences before and after matching.

Court-Mandated Family Violence Interventions Central Connecticut State University Figure 8. Line Plot of Standardized Matching Differences Figure 9 below shows histograms with overlaid kernel density estimates of standardized differences before and after matching.

Figure 9. Histograms of Standardized Matching Differences Comparison of EVOLVE Study Groups Following the matching process it was necessary to assure the two study groups were equal in terms of race/ethnicity, DVSI-R assessment scores, age, and criminal history.

There were minimal percentage differences in the racial/ethnic make-up of the two study groups and Court-Mandated Family Violence Interventions Central Connecticut State University these differences were not statistically significant (Table 23). African-Americans comprised the highest percentage of EVOLVE participants (37%) compared to Hispanics (32%) and whites (29%).

–  –  –

Table 24 presents the comparison of DVSI-R assessment scores between the two study groups with no statistically significant differences. For the DVSI-R risk score categories, offenders in the EVOLVE treatment group and the EVOLVE comparison group were mostly high risk (41% and 40% respectfully) or very high risk (33% for the treatment group and 34% for the comparison group). The majority of offenders in both groups had posed a high risk to the victim (77% for the EVOLVE comparison group and 74% for the EVOLVE treatment group).

–  –  –

The final comparative analyses explored average differences in offenders’ age at the time of arrest, their DVSI-R total risk scores, number of prior arrests, and number of prior family violence arrests (Table 25). There were no statistically significant differences between the two study groups for any of these factors. The average age for both study groups was 31 years old, their DVSI-R total risk score was 14, and both groups had an average of eight prior arrests and about two prior family violence arrests.

Court-Mandated Family Violence Interventions Central Connecticut State University

–  –  –

Differences Between EVOLVE Program Completers and Non-Completers The EVOLVE program had a completion rate of 65% (120 out of 185 participants were successfully discharged). In looking at the completion rate by race/ethnicity, Hispanic participants had the highest completion rate (78%) followed by white (68%) and AfricanAmericans participants (51%)(Table 26). These percentages were much different than for EXPLORE. The EXPLORE completion rates were 51% of Hispanic and 70% of AfricanAmerican participants.

–  –  –

There were also statistically significant differences across DVSI-R risk categories for EVOLVE completers and non-completers (Table 27). Very high risk offenders were much less likely to complete EVOLVE (52%) than low risk (80%), moderate risk (63%), or high risk offenders (76%). There were no differences in the risk to victim measures, however, this was likely due to the low number of EVOLVE participants in the medium risk (42) and low risk (7) categories.

Court-Mandated Family Violence Interventions Central Connecticut State University

–  –  –

Table 28 shows there were no statistically significant differences between EVOLVE completers and non-completers for age, DVSI-R total risk score, prior number of arrests, and prior number of family violence arrests.

–  –  –

EVOLVE One Year Arrests The following tables present the follow-up outcome analyses of one year arrest rates for any new arrest and for new family violence arrests. For any new arrest, the EVOLVE treatment group had a much lower one year arrest rate (35%) than the EVOLVE comparison group (55%)(Table 29). These differences were statistically significant.

–  –  –

While the EVOLVE treatment group was less likely to be arrested than the EVOLVE comparison group for new arrests, these differences were not present for new family violence arrests (Table 30). The EVOLVE treatment group did have a lower one year arrest rate (22%) than the EVOLVE comparison group (29%), however, these differences were not statistically significant.19

–  –  –

EVOLVE Effect Size Calculations The effect size statistics for EVOLVE are presented in Table 31. The effects are moderate for any new arrest (-0.50) and small but not statistically significant for new family violence arrests (-0.22). Similar to the other effect sizes, the lower the effect size the more effect EVOLVE had in lowering the recidivism of program participants. The odds ratio demonstrates the magnitude of the effects, in that, offenders who were eligible for EVOLVE but did not participate were 2.27 times more likely to be arrested for any new offense than EVOLVE participants.

–  –  –

Note: It is quite possible that the small sample size for the EVOLVE study group can be a factor explaining the lack of significance for these differences.

Court-Mandated Family Violence Interventions Central Connecticut State University

–  –  –

The study assessed the effects of three-court mandated family violence programs overseen by the Judicial Branch’s Court Support Services Division. One of these programs, the Family Violence Education Program (FVEP), is a 9-session pre-trial program for male and female family violence offenders. The other two programs, EXPLORE and EVOLVE, are postplea programs specifically targeting male batterers. EXPLORE is a 26-session program (groups meet once a week for 90 minutes) while EVOLVE is more intensive and meets for 52 sessions (groups meet for 2 hours twice a week for 26 weeks).

This study was mandated by Connecticut Public Act 13-247 An Act Implementing Provisions of the State Budget to provide effect sizes for each program to be used in the development of criminal justice cost-benefit analysis models as part of the General Assembly’s Results First Initiative. As such, this study centered on the effects each program had on subsequent criminal behavior.

The evaluation was conducted using a quasi-experimental research design with propensity-matched comparison groups. While ideally programs should be evaluated by studying people who are randomly assigned to participate in a treatment program or assigned to a nontreatment control group, such a design was not practical given the time constraints of the study and the legal and ethical concerns associated with denying treatment to serious and potentially violent offenders. The matching process consisted of identifying similar offenders who were eligible for each program but did not attend. We believe the matching process was successful in creating comparison groups closely related to offenders in each program.

Summary of Findings

The study had three research questions: (1) what were the completion rates for each program and were there statistically significant differences between program completers and non-completers; (2) was the one-year arrest rates for new offenses or family violence offenses of offenders who participated in the program statistically significantly different from those offenders who did not participate in the program; and, (3) were there measureable program effect sizes?

Pages:     | 1 |   ...   | 5 | 6 || 8 | 9 |   ...   | 11 |

Similar works:

«FORWARD Please note – this handbook can be found online at: http://www.saintleo.edu/international-services This handbook is designed to be used with the Saint Leo University Student Handbook. All efforts have been made in this handbook to provide useful and current information. If this handbook does not address an issue, please feel free to contact any of the professional staff members below: Lawson Jolly, LMHC # MH4058 Director of Counseling Services deChantal Hall, Office # 123, (352)...»

«    Faculty Publications  UC Hastings College of the Law Library    Author:  Ben Depoorter    Source:  Cornell Law Review  Citation:  95 CORNELL L. REV. 957 (2010).  Title:  Law in the Shadow of Bargaining: The Feedback Effect of Civil Settlements        Originally published in CORNELL LAW REVIEW. This article is reprinted with permission from CORNELL  LAW REVIEW and Cornell University Law School.  ESSAY LAW IN THE SHADOW OF...»

«OIPC File No: F15-63155 Inquiry under Part 5 of the Freedom of Information and Protection of Privacy Act (FOIPPA) between: An Applicant and the Vancouver Police Department (VPD) and BC Association of Chiefs of Police (intervenor) and BC Association of Municipal Chiefs of Police (intervenor) and BC Civil Liberties Association (intervenor) and BC Freedom of Information and Privacy Association (intervenor) and Open Media (intervenor) and Canada Research Chair of Ethics, Law and Technology,...»

«ENFORCING EQUAL EMPLOYMENT OPPORTUNITIES IN CHINA XUN ZENG I. INTRODUCTION: MR. JIANG TAO'S STORY' In December 2001, Jiang Tao, a law student at Sichuan University, read an advertisement, which indicated that the Chengdu Branch of People's Bank of China (PBOC), was hiring new staff members. The advertisement contained a minimum height requirement: male candidates were required to be at least 168cm (5ft 6in) tall and females 155cm (5ft 3in).3 Such height requirements are very common in China;...»

«Republic of Macedonia Government of the Republic of Macedonia Law on Territorial Organization of the Local Self-Government in the Republic of Macedonia 11 December 2003 CONTENTS of the Law on Territorial Organization of the Local Self-Government in the Republic of Macedonia I. GENERAL PROVISIONS Article 1 This law regulates the territorial organisation of the local self-government in the Republic of Macedonia, establishes the municipalities, determines the territory of the municipalities and...»

«A copy of this preliminary short form prospectus has been filed with the securities regulatory authorities in each of the provinces in Canada, other than the Province of Québec, but has not yet become final for the purpose of the sale of the securities. Information contained in this preliminary short form prospectus may not be complete and may have to be amended. The securities may not be sold until a receipt for this short form prospectus is obtained from the securities regulatory...»

«1 POLICING ACROSS BORDERS PROJECT  Third Workshop STRENGTHENING COOPERATION IN THE FIGHT AGAINST DRUG TRAFFICKING: LEGISLATION, INSTITUTIONS, AND PROPOSALS Bio Sketches Ms. Ela Banaj is a National Project Officer, in Albania, of the United Nations Office on Drugs and Crime (UNODC) since April 1998. Ms. Banaj holds a degree in English Language and Literature, a degree in law from the University of Tirana and a MA Degree. Her Master’s thesis was on “Drug-related Penal Offences in the...»

«VILNIUS UNIVERSITY VIOLETA VASILIAUSKIENĖ THE FIGHT AGAINST TERRORISM IN THE CONTEXT OF INTERNATIONAL HUMANITARIAN LAW Summary of Doctoral Disertation Social Sciences, Law (01 S) Vilnius, 2014 Dissertation was written in 2007 – 2014 at Vilnius University.Scientific leader: Prof. dr. Dainius Žalimas (Vilnius University, Social Sciences, Law – 01 S). Dissertation is defended at the Law Science Council of Vilnius University: Chairman: Prof. dr. Armanas Abramavičius (Vilnius University,...»

«National District Attorneys Association American Prosecutors Research Institute 99 Canal Center Plaza, Suite 510 Alexandria,VA 22314 www.ndaa.org Thomas J. Charron President Roger Floren Chief of Staff Jennifer Gentile Long Director, National Center for the Prosecution of Violence Against Women David LaBahn Director, NDAA Research and Development Division 2007 by the American Prosecutors Research Institute, the research and development division of the National District Attorneys Association....»

«COLLEGE OF MIDWIVES OF BRITISH COLUMBIA MIDWIFERY MODEL OF PRACTICE DEFINITION OF A MIDWIFE A midwife is a person who, having been regularly admitted to a midwifery educational programme duly recognised in the country in which it is located, has successfully completed the prescribed course of studies in midwifery and has acquired the requisite qualifications to be registered and/or legally licensed to practise midwifery. International Definition of the Midwife. SCOPE OF PRACTICE OF THE MIDWIFE...»

«ABORIGINAL PEOPLE AND THE MEDIA REPORTING ABORIGINAL AFFAIRS Diana Plater IN LATE 1985 I WAS AUSTRALIAN ASSOCIATED PRESS'S WESTERN Australian correspondent. While I was in Kununurra, the family of a young Aboriginal man who had died in Broome gaol came to visit. They were desperate. They did not believe the cause of death suggested by the prison and suspected some foul play. For this reason they wanted an independent post mortem, only they could not afford it. They had asked the local legal...»

«Journal of Information Technology Education: Volume 10, 2011 Innovations in Practice Using Digital Logs to Reduce Academic Misdemeanour by Students in Digital Forensic Assessments Harjinder Singh Lallie Philip Lawson University of Warwick, University of Derby, School of International Digital Computing and Mathematics, Laboratory, Coventry, UK Derby, UK h.s.lallie@warwick.ac.uk p.lawson@derby.ac.uk David J. Day Sheffield Hallam University, Sheffield, UK d.day@shu.ac.uk Executive Summary...»

<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.