DVHF Demonstration Evaluation

Name: DVHF Demonstration Evaluation
Published: 2023-04-02
License: https://creativecommons.org/licenses/by/4.0

Name	File Type	Size	Last Modified
DVHF Advocate Data.sav	application/x-spss-sav	45.7 KB	04/02/2023 06:37:AM
DVHF Agency Data.sav	application/x-spss-sav	13.8 KB	09/27/2022 07:39:AM
DVHF Participant Data 12mo.sav	application/x-spss-sav	542.7 KB	09/27/2022 07:42:AM
DVHF Participant Data 18mo.sav	application/x-spss-sav	639.8 KB	09/27/2022 07:42:AM
DVHF Participant Data 24mo.sav	application/x-spss-sav	467.1 KB	09/28/2022 01:06:PM
DVHF Participant Data 6mo.sav	application/x-spss-sav	678.5 KB	09/27/2022 07:36:AM
DVHF Participant Data Baseline.sav	application/x-spss-sav	655.6 KB	09/27/2022 07:42:AM

Project Citation:

Sullivan, Cris M. DVHF Demonstration Evaluation. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2023-04-02. https://doi.org/10.3886/E187961V1

Project Description

Summary: This longitudinal, quasi-experimental study evaluated the impact of the Domestic Violence Housing First (DVHF) model on the safety, housing stability, and well-being of domestic violence survivors over the course of two years. DVHF includes providing survivors with housing-inclusive advocacy and/or flexible funding. Four hundred six homeless or unstably housed DV survivors were interviewed shortly after contacting services for help. They were then interviewed every six months over two years. Retention was 89% at 24-months. Evidence indicates that the DVHF model is more effective than services as usual (SAU) in helping survivors achieve housing stability, safety, and improved mental health over twenty-four months. Survivors who received DVHF also reported higher prosocial behaviors from their children compared to parents who received SAU. Positive change in these domains happened quickly (within the first 6 months after seeking services) and persisted across 12, 18, and 24 months. The model does not appear to be more effective than SAU in increasing financial stability or reducing substance misuse. It also showed no impact on children’s school attendance, school performance, nor on their behavioral problems. The DVHF model worked similarly across people from various race and ethnicity groups, as well as both urban and rural geographic service areas. For participants who had received DVHF, the extent to which they reported agencies engaging in trauma-informed practices was positively related to their housing stability, safety, and mental health over 24-months follow-up.

Funding Sources: This research was supported by a subcontract from the Washington State Coalition Against Domestic Violence, who received funding through a contract with the U.S. Department of Health and Human Services’ Office of the Assistant Secretary for Planning and E

Scope of Project

Subject Terms: intimate partner violence; housing; homelessness; intervention

Geographic Coverage: Washington state

Time Period(s): 2017 – 2022

Collection Date(s): 2017 – 2022

Universe: Adult survivors of recent intimate partner violence who also self-identified as being homeless or unstably housed.

Data Type(s): administrative records data; survey data

Collection Notes: Interviews with survivors were conducted in person or by phone.

Methodology

Response Rate: Retention at 6-months was 92% (n=375). At 12-months it was 91% (n=369), at 18-months it was 88% (n=359) and at 24-months it was 89% (n=363).

Sampling: Participants were recruited from five DV organizations (two urban, three rural) in a state in the Pacific Northwest. Each agency confirmed that they subscribe to the DVHF model but that there are times they do not have the funding nor staffing to provide housing-focused, mobile advocacy. During the time of study recruitment, DV agency staff agreed to inform all new clients who were homeless or unstably housed about the study. They referred 597 clients who were interested in hearing more about the study. We successfully reached 514 of these clients (86%) and told them more about the study. Fifteen percent were ineligible for the study because they either had not experienced recent IPV or were not either homeless or unstably housed. Seven percent declined to participate after hearing more (eight participants specifically noting safety concerns). The final sample consisted of 406 participants (93% of the 438 eligible clients). While agency staff kept no written documentation to support this, they verbally confirmed that few clients were ineligible to participate in the study and that those who enrolled in the study were similar demographically to all their clientele.

Data Source: In-person or phone interviews with survivors; agency records of services and funding provided; one web-based survey of advocates at 6-months

Collection Mode(s): computer-assisted personal interview (CAPI); face-to-face interview; record abstracts; telephone interview; web-based survey

Scales: Composite Abuse Scale
Scale of Economic Abuse Revised
Use of Children to Control
Housing Instability Scale
Financial Strain subscale of Scale of Economic Hardship
Inability to Make Ends Meet subscale of Scale of Economic Hardship
Herth Hope Index
Quality of Life Scale
Patient Health Questionnaire 9 (PHQ-9)
Generalized Anxiety Disorder 7 (GAD-7)
Trauma Screening Questionnaire
Medical Outcomes Study Social Support Survey (MOS-SSS-6)
CAGE-AID
Modified Differential Emotions Scale
Measure of Victim Empowerment Related to Safety (MOVERS)
Child Strengths and Difficulties Questionnaire

Weights: Inverse-probability-weighted (IPW) estimators were calculated and included in the longitudinal models as sampling weights to account for any selection bias resulting from factors that increased the probability that certain individuals received the intervention. To compute the IPW estimators, we first used logistic regression models to examine whether there were any meaningful baseline differences between those who received DVHF versus those receiving SAU. Thirteen factors were identified and used to calculate the IPW estimator: parenting children (y/n), living with the abuser (y/n), racial/ethnic minority (y/n), having been in foster care as a child (y/n), housing barriers, housing instability, staying with friends to avoid homelessness (y/n), inability to make ends meet, overall abuse, alcohol misuse, drug misuse, quality of life, and whether or not the DV agency was in a rural area (y/n).

Unit(s) of Observation: Individuals

Geographic Unit: N/A

Related Publications

Download this project

Published Versions

V1 [2023-04-02]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.