Violence Against Women and the Role of Welfare Reform in Stanislaus and Kern Counties, California, 1999-2002 (ICPSR 3797)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Hide

Study Purpose View help for Study Purpose

This study investigated the relationship between domestic violence and welfare reform. While a number of studies prior to the enactment of welfare reform demonstrated that high rates of welfare recipients experienced domestic violence, this study was designed to provide comprehensive information about domestic violence under the new welfare rules. Two specific goals were (1) to determine the impact of domestic violence on welfare tenure and employment over a three-year period, and (2) to examine the well-being of children of female welfare recipients who experienced domestic violence. In addition, the study examined issues related to mental health problems and alcohol and drug problems. Both conditions are associated with domestic violence, but the extent of this link among welfare recipients after welfare reform was enacted was unclear.

Study Design View help for Study Design

This study was based on three rounds of interviews with a random sample of welfare recipients in Kern and Stanislaus Counties, California, between April 1999 and December 2002. Interviews were conducted at the welfare department in each county and were intended to occur on days when participants had other, already scheduled activities there. As an incentive and compensation for time and travel, study participants in Round One were offered a 30 dollar gift card for Wal-Mart. The gift card was increased to 50 dollars in Round Two and 75 dollars in Round Three. There turned out to be many problems with this approach. Participants were often not at the site at the time at which they were scheduled, requiring the interviewers to try to contact them by letter and telephone to come in for the interview. Home visits were not part of the study design, primarily to protect the safety of women who might be in abusive relationships. The first round of interviews was conducted from March to October 1999 and involved 356 subjects from Stanislaus County and 276 subjects from Kern County. Round Two interviews were conducted from May to December 2000 and involved 311 subjects from Stanislaus County and 262 subjects from Kern County. Round Three interviews took place between January 2001 and December 2002 and involved 309 subjects in Stanislaus County and 243 subjects in Kern County. The interview instrument was slightly different for each round. In all three rounds the questions focused on employment history, health, mental health, domestic violence victimization, children's well-being, and alcohol and drug use. The instruments for Rounds Two and Three included additional items on children's well-being, and the Round Three instrument included a section on criminal history. The instruments were a synthesis of the following instruments along with items designed by the researchers: the Composite International Diagnostic Interview (CIDI), the Composite International Diagnostic Interview Short Form (CIDI-SF), the Short Form 12 (SF-12) and Short Form 36 (SF-36) Health Surveys, the Rosenberg Self-Esteem Scale, the Behavior and Symptom Identification Scale (BASIS-32), and the Ohio Youth Problems and Functioning Scales (only in the Round Three instrument).

Sample View help for Sample

Kern and Stanislaus counties were chosen as sites for this study because of their leadership in developing ideas for working with welfare recipients and their emphasis on cooperative planning between local domestic violence centers and mental health/substance abuse and welfare departments. Sampling criteria included being eligibile for welfare, being the head of the household (not in a two-parent family) and the mother (not another relative) of any children living in the household, fluent in English or Spanish, between the ages of 18 and 59, and receiving welfare for at least one year at the time of the first interview (if in Kern County) or applying for welfare (in Stanislaus County). For logistical reasons the Kern County sample was limited to Bakersfield, the largest city in Kern County. The Kern County sample was selected from a list of 4,732 welfare recipients generated by the Kern County Department of Human Services who met the above criteria and who were recertified to receive welfare between April 1, 1999, and August 30, 1999. In Kern County a stratified random sample was drawn for each month during the sampling period, with stratification on age (over 25 or not) and duration of time on welfare (two years and over or less). The Stanislaus County sample was drawn from women who applied for welfare in Stanislaus County between May 1, 1999, and August 15, 1999.

Universe View help for Universe

Women on welfare in Kern County whose recertification dates fell between April 1, 1999, and August 30, 1999. Women who applied for welfare in Stanislaus County between May 1, 1999, and August 15, 1999.

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Description of Variables View help for Description of Variables

Variables in Part 1 include eligibility for welfare during each round of interviews, whether cash aid was received by respondent, education, age, race, marital status, number of children, childcare situation, employment history, other sources of financial support, quality of current housing, health questions from the SF-12 and SF-36, domestic violence victimization history, abuse endured/witnessed in childhood, questions from the CIDI, section K, related to post-traumatic stress disorder, questions from the CIDI-SF, which included sections on major depressive episodes, generalized anxiety disorder, social phobia, agoraphobia, other specific phobias, and panic attacks, Diagnostic and Statistical Manual, Fourth Edition scales derived using the answers to the CIDI-SF questions, questions and scales from the BASIS-32, questions and score from the Rosenberg Self-Esteem Scale, history of mental health treatment, questions on history of alcohol and drug use and treatment taken from the CIDI sections J and L, and many derived variables. The variables in Part 2 are the same as in Part 1, with additional variables on child well-being, housing and material hardship, childcare and transportation, experiences with the California welfare-to-work program, social support, income and welfare benefits, and health insurance for children. Variables on alcohol and drug use in Part 2 correspond to questions from the CIDI-SF and were used to calculate Diagnostic and Statistical Manual, Third Edition, Revised scales. CIDI sections J and L are not in Part 2. There are many derived variables including several related to child well-being that are not in Part 1. The variables in Part 3 are the same as in Part 2 except the Ohio Youth Problems and Functioning Scales were added to the Round Three instrument along with other items related to child well-being and criminal history. In addition, Part 3 does not contain all of the derived variables that Part 2 does.

Response Rates View help for Response Rates

Of the Stanislaus study-eligible applicants 71 percent were interviewed in Round One. In Kern County 55 percent of the eligible sample was interviewed in Round One. Of the 276 Kern respondents in Round One 262 (95 percent) were interviewed in Round Two, and 243 (88 percent) were interviewed in Round Three. Of the 356 women interviewed in Round One in Stanislaus County, 311 (87 percent) were interviewed in Round Two, and 309 (87 percent) were interviewed in Round Three.

Presence of Common Scales View help for Presence of Common Scales

Scales include the Composite International Diagnostic Interview, the Composite International Diagnostic Interview Short Form, Diagnostic and Statistical Manual, Third Edition, Revised, Diagnostic and Statistical Manual, Fourth Edition, the Short Form 12 and Short Form 36 Health Surveys, the Rosenberg Self-Esteem Scale, the Behavior and Symptom Identification Scale, and the Ohio Youth Problems and Functioning Scales.

Hide

Original Release Date View help for Original Release Date

2005-02-25

Version History View help for Version History

Hide