The study was designed to:
- Empirically describe perceived barriers to help-seeking for female victims of domestic abuse age 50 years and older who were not in the service system for abused women relative to the perceived barriers for women in the same age group who are not victims of such abuse.
- Empirically describe the impact of key demographic variables of race and ethnicity, relationship to the presumed abuser, age at the time of the survey and type of abuse on perceived barriers of victims.
- Collaboratively develop a draft coordinated community response (CCR) plan.
Subjects were recruited primarily via ads in newspapers, flyers and announcements provided at senior centers. At two points during subject recruitment, which was ongoing for almost one year, an announcement was sent out to a mailing list of local agencies and organizations that were likely to have elder constituents, offering to send flyers if there was an appropriate audience. At the conclusion of each session flyers were distributed to study participants to share with friends, neighbors, and associates in any groups to which they belonged.
The project design called for data to be collected using a self-administered questionnaire (Quantitative Survey Data) in a small group setting. After questionnaires were completed and collected, a brief discussion was held regarding issues covered in the survey questions and help-seeking options were described. Each participant received a copy of a booklet (available in English and Spanish) that offered information about domestic abuse in later life and the broader issue of elder abuse, and described community resources for women who wanted assistance. All participants who attended a data collection session received a $25 cash stipend for their participation, regardless of whether they were able to complete the questionnaire.
Twelve open-ended interviews (Qualitative Interview Data) were conducted as a follow-up to the survey questionnaire. All of the respondents who participated in the follow-up interviews had responded "yes" to survey item 77, which asked respondents if they had experienced any of the problems described in the questionnaire after age 50. All had signed a consent form confirming their willingness to be contacted for an interview. The interviews were conducted in either English or Spanish, according to each participant's preference. Respondents first were asked to describe their relationship to their abuser and then presented with the questions from each section of the PBHS Assessment, i.e., behavior of the abuser, internal barriers to help-seeking and external barriers to help seeking. Respondents were asked to clarify and discuss their responses in their own words. Each respondent was encouraged to talk in some detail about whether they would chose (or had chosen) to discuss the abuse they had experienced with others. They were also offered an opportunity to discuss anything that they felt was important but not covered in their previous responses. Interviews were audio recorded and transcribed verbatim by the interviewer. The interview transcripts were entered into Atlas.ti. Transcripts were coded in their original language using a combination of a priori codes that followed the sections of the quantitative questionnaire and open coding.
The recruitment strategy focused on attracting survey respondents who were evenly represented in the three targeted racial-ethnic subgroups: Hispanic, White-non-Hispanic, and Black-non-Hispanic (150 in each group). Initially minimum age of 50 years was the only criterion for screening potential subjects. Once the 150 completed surveys target for a particular racial-ethnic subgroup was reached, then screening on race-ethnicity for the other racial-ethnic groups continued until the target was reached for each group.
Ads and flyers used the description "survey about conflict in relationships" for older women. A total of 519 participants signed consent forms and received a stipend for some level of participation. An additional 186 women participated in three pilot studies (48 in each of the first two pilots and 90 women in the third) to test the readability and distribution assumption of the instrument, and also received a stipend. Subjects were recruited primarily via ads in newspapers, flyers and announcements provided at senior centers. At two points during subject recruitment, which was ongoing for almost one year, an announcement was sent out to a mailing list of local agencies and organizations that were likely to have elder constituents, offering to send flyers if there was an appropriate audience. At the conclusion of each session flyers were distributed to study participants to share with friends, neighbors, and associates in any groups to which they belonged.
Preliminary analyses were conducted on data from the 486 surveys with an MSQ score of 8, 9 or 10, which indicated a level of cognitive ability adequate to function independently in the community and to complete the questionnaire. Cases were dropped for participants that did not provide sufficient data to obtain an estimate of an indicator or factor. Application of these filters resulted in exclusion of 41 surveys, leaving data for 445 participants.
A total of 445 women age 50 and older were recruited and met criteria for inclusion of their survey data in the analyses: (1) sufficient cognitive functioning to complete the questionnaire; and (2) completed a sufficient number of survey items required for statistical analyses. All 445 participants were not in the service system for abused women. Twelve women who self-identified as being victims of abuse after they were age 50 volunteered to participate in an interview regarding their experiences and barriers. Twelve professionals representing seven agencies participated in an advisory group.
Females age 50 years or older interested in participating in research regarding conflict in close personal relationships experienced by women in this target age range living in Miami-Dade county.
The study contains 116 variables including described abuser behaviors, personal attitudes and beliefs, experiences of domestic abuse, mental status/cognitive impairment, and demographic characteristics. Described abuser behaviors information included questions about whether the respondent was afraid of the abuser, whether the abuser gets angry with the respondent if they talk to anyone the abuser does not know, whether the abuser makes it difficult for the respondent to spend time with their family, whether their decisions are based on what the abuser wants them to do, whether the abuser suspects that the respondent is cheating on them, whether the respondent has no privacy, and whether the abuser would hurt the respondent if they told anyone about their problems. Personal attitudes and beliefs information included whether the respondent thought a women her age should tell someone if she was hurt by a person close to her, take care of her family no matter how they treat her, is responsible for making the people she is close to happy, should do whatever it takes to keep her family together, put the needs of her family before her own, cannot change problems in her close personal relationships, should discuss family problems with a doctor or counselor, whether the respondent thought there is nothing a woman her age can do about emotional or physical abuse when it has been going on for many years, and whether the respondent thought that it is better for a woman her age to be with someone who treats her badly than for her to be alone. In addition the personal attitudes and beliefs variables asked the respondent whether a woman her age should tolerate being scared or hurt by her husband, according to their religious beliefs, whether family members expect a woman her age to keep the family together no matter how she is treated, whether she thought the police are helpful to women her age who are hurt by people close to them, whether friends expect a woman her age to "stick it out" when they are hurt by someone close to them, whether she thought that police will not do anything to help when they go to a house where there is domestic abuse involving a woman her age, whether she thought that family members would get angry at a woman her age who presses charges with the police when someone close to her hurts her, whether she thought that the community she lives in has services to help a woman her age who is hurt by a family member or close friend, and whether she believes a woman her age is to blame when she is hurt by another family member or close friend. Experiences of domestic abuse were gathered using the Conflict Tactics Scales-Revised, Short Version (CTS2S). This scale measured whether the respondent experienced domestic abuse by asking the respondent how often their abuser insulted or swore at them, suffered a sprain, bruise, or small cut or felt pain the next day because of a fight with their abuser, showed respect for, or showed that they cared about the respondent's feelings about an issue they disagreed on, punched, kicked or beat up the respondent, destroyed something belonging to the respondent or threatened to hit them, or used force to make the respondent have sex with them. Mental status/cognitive impairment information was collected by asking respondents what was the name of the building they were in at the time, where the building was located, what day of the month it was, what day of the week it was, what year it was, how old they were, when their birthday is, what year they were born, what was the name of the President at the time, and who the President was before the current one. Demographic characteristics included race, age, marital status, and gender.
A total of 519 subjects were recruited to participate in the study. Of these 519 participants, a total of 445 women age 50 and older met criteria for inclusion of their survey data in the analyses.
Conflict Tactics Scales-Revised, Short Version (CTS2S) - Ten item scale used to determine if a participant had experienced one or more forms of domestic abuse, including psychological abuse, physical abuse, sexual coercion, and injuries resulting from the abuse.
Several other Likert-type scales were used.