RELOADED Disability Studies Quarterly Fall 2007, Volume 27, No.4 <www.dsq-sds.org> Copyright 2007 by the Society for Disability Studies

In a stratified society characterized by inequality, certain groups experience marginalization. Marginalization refers to the unequal distribution of resources and power which produces conflict between groups, leaving some more vulnerable to abuse than others. Persons with disabilities are one such group (Oliver, 1996; Campbell and Oliver, 1996). They are marginalized socially, economically and politically and frequently are treated as a subordinate group within society. Individually and collectively they are denied full participation in society. Their marginalization places them at a serious disadvantage within the community. People with disabilities experience conflict in a variety of forms, including problems of physical accessibility, lack of employment opportunities and lower salaries, and problems accessing health care.

One important manifestation of the marginalization of persons with disabilities can be seen in the rate of abuse directed at people with disabilities. Research on abuse traditionally defined as physical, emotional, sexual and financial abuse and neglect has documented that people with disabilities experience rates much higher than the rate for people without disabilities (Crosse et al., 1993; Sobsey et al., 1995; Sullivan and Knutson, 2000; Wilson and Brewer, 1992). Traditional methods of capturing data on crime victims, however, such as the Bureau of Justice Statistics National Crime Victimization Survey and the Federal Bureau of Investigation's Uniform Crime Report have not included data on the crime victim's disability status.

While abuse in the literature is defined by researchers and providers as physical, sexual, emotional, and financial abuse and neglect, there are very few studies which focus on how people with disabilities actually define abuse. The Roeher Institute (1995) surveyed a sample of people with disabilities (as well as conducted interviews and focus groups with service providers, police, advocates, and family members). They found that types of violence and abuse identified by persons with disabilities included: physical violence; denial of rights, necessities, privileges, and opportunities; verbal and emotional abuse; and failure to respond to their complaints of abuse and violence. In addition, there have been studies of other marginalized populations where it was found that members of the affected population have defined abuse more broadly. For example, Lewis, et al (2005) studied intimate partner violence in the Latino community and found that women (many of whom were victims of violence) added another category to possible types of abuse. In addition to identifying traditional types of abuse (physical, sexual, emotional, and financial abuse), they spoke of the unequal burden of expectations imposed on women in society as a type of abuse. This abuse (identified as "unequal burden") referred to the abusive expectation that women were not only expected to cook, clean house, shop and prepare food, care for children and aging parents, and meet the needs of their spouse, but they were also expected to contribute financially as a breadwinner within the family — and in many cases, the only breadwinner.

A similar understanding of how people with disabilities actually define abuse is missing from the literature. The overall goal of our research was to develop a better understanding of abuse directed at people with disabilities — from their perspective. 01 In particular, we wanted to learn about types of behaviors identified as abusive, likely abusers, reporting behaviors and perceived outcomes. This paper focuses on the way in which abuse is defined by people with disabilities and who they identify as likely abusers. Findings with regard to reporting abuse have been presented elsewhere (West and Gandhi, 2005).

Methodology

Focus groups to discuss the issue of abuse were conducted with people with disabilities who were a part of a self-advocacy or support group. While some may take issue with the fact that we were asking people to discuss a very sensitive issue using a public format, we felt that engendering a general discussion in a self-support group where participants felt supported by one another as they offered ideas was an appropriate methodology to use as a first step in the process of exploring this issue with people with disabilities. According to Krueger (1988) a focus group is

a carefully planned discussion designed to obtain perceptions on a defined area of interest in a permissive, non-threatening environment. It is conducted with approximately seven to ten participants by a skilled interviewer. The discussion is relaxed, comfortable, and often enjoyable for participants as they share their ideas and perceptions. Group members influence each other by responding to ideas and comments in the discussion.

The focus group approach was selected as an appropriate method to explore a sensitive topic with people with disabilities. The introduction by the moderator was designed to make people comfortable in sharing their views. It is our intention to continue our research in this area and next explore this issue in one-on-one interviews with people who have self-identified themselves as victims of abuse.

Six focus groups were conducted in 2004. The purpose of these groups was to explore the issue of abuse directed at people with disabilities. With help from two self-advocacy organizations ("Monday Morning Networks" organized by the NJ Council on Developmental Disabilities [NJCDD] and the NJ Self-Advocacy Project of The Arc of New Jersey) participants were recruited for the focus groups; both organizations represent a network of smaller organizations that meet on a regular basis around the state. Membership is limited to adult consumers ages 18 and older with a wide range of disabilities, including physical and cognitive disabilities. Some members have developmental disabilities; others have become disabled later in life. While some members lived in institutional settings in the past, all are currently living within the community either independently or in supervised living settings.

Rather than recruiting from the general population of persons with disabilities, recruitment was limited to members of these self-advocacy groups. It is important to remember that these individuals may be unique because of their participation in outside groups. We worked through self-advocacy groups whose members have received self-assertiveness training and workshops on public speaking and who frequently participate in public forums and presentations because we hoped members would be more willing to talk about the issue of abuse.

With the help of several members of the New Jersey Self-Advocacy project, focus group questions were drafted, pre-tested and then revised. Focus groups were held in settings around New Jersey in the north, central and southern regions of the state. Groups were facilitated by the principal investigator and co-facilitated by a graduate student. In addition, two members of the New Jersey Self-Advocacy project were trained as co-facilitators. The presence of the self-advocate serving as a co-facilitator helped ease entry into groups. The individuals helped answer participants' questions, suggested probes and made participants feel more comfortable during discussions.

All focus groups were held in physically accessible locations including public library meeting rooms, the offices of disability organizations, and city municipal buildings. These locations were settings where the self-advocacy groups were meeting on a regular basis. In most cases, transportation was prearranged by the self-advocacy groups; however, where assistance with transportation was needed, arrangements were made.

At the start of each meeting, the purpose of the study was explained and persons who agreed to participate were then asked to initial a consent form. An alternative activity was made available for anyone who did not want to be a part of the focus group. On average, focus groups were one and one-half hours in length. At the conclusion of the focus group, a small incentive was provided to each participant to thank them for their participation.

In starting the focus groups, the facilitator explained that the purpose of the discussion was to learn more about "seriously wrong things some people do to people with disabilities. These are things considered wrong by the person with a disability, including seriously wrong things people do that physically hurt a person or scare them, or that take away things that belong to the person." This definition was developed through discussions with staff at The Arc of New Jersey and with the advice of our co-facilitators.

Once the definition of abuse had been provided, the facilitator then asked participants to offer examples of things people do to people with disabilities that are wrong or bad. Facilitators explained to participants that they did not have to talk about anything they were not comfortable discussing in the group. When examples were provided, the facilitator did not ask the individual if the situation was one they had personally experienced or whether it involved anyone they knew. Counselors were on hand at the meetings to follow-up with situations where people wanted to talk further after the focus group. Only in one case was a counselor called upon to discuss a situation that came up during a focus group.

Question Areas

Talking about abuse:

• Give me some examples of things people do to people with disabilities that you think are wrong or bad.
• Who is likely to do these things to people with disabilities?

After the participants stopped providing examples, participants were then asked about other kinds of abuse that had not been mentioned. For example, if no examples of financial abuse were provided by participants, the facilitator would then ask about this type of abuse. In coding the transcripts, examples "first mentioned" without probes were coded as FM and those offered "after probes" were codes as AP. Participants were also asked about abusers — who were the people likely to do these things to people with disabilities. They were also asked about settings where abuse was more likely to happen, whether people with disabilities report these types of situations when they happen, why or why not, and what happens when they do report. Each focus group ended with a general discussion of things people can do to be safe from abuse. As noted above, data from only the first two questions are reported here. Findings from the other questions have been reported elsewhere (West and Gandhi, 2006).

All focus group sessions were audiotape recorded with the permission of participants and tapes were transcribed. A content analysis of the major themes in the transcripts was then undertaken using Atlas.ti (Version 5.0) — a software program that helps with the process of coding long narratives for themes. Once coded, the findings can be sorted by theme. A group of initial codes were created based on major themes found in the literature and then supplemented with additional themes drawn from reading the transcripts. The themes were then applied to code each transcript in Atlas.ti. The coding completed by a graduate student was then compared with coding done by the principal investigator to assure agreement in the application of codes and differences were discussed and resolved.

Based on a reading of the transcripts, types of abuse were coded as physical, emotional, sexual, financial, neglect, denial of rights, isolation and other. Certain examples were offered that could have been cross categorized in several ways, in particular, some examples of neglect and emotional abuse could be coded into either category. Several situations were described by participants that could have been categorized as either neglect or emotional abuse. In these situations, the two coders reviewed the full transcript and discussed the situation. It was agreed that if the person described the situation in terms of its emotional impact on them, it would be coded as emotional abuse. If it was offered with no reference to any emotional impact on them, it was coded as neglect. In describing a situation, if a participant actually used a term being used for categorizing a type of abuse such as "physical abuse," "neglect," or "isolation," we coded the example accordingly — even if the example seemed more suitable under another category of abuse. For example, in one situation, a participant described a situation as "neglect" — even though it could have been coded as "isolation." We coded the example according to the participant's characterization.

Abusers were coded as either unknown or known. Known abusers were further differentiated between "known intimates" who included family and partners and "known acquaintances" who included friends, institutional staff, community caregivers, students/teachers, health professionals, coworkers/boss, police/security personnel, or other people with disabilities. In several examples, a participant referred to an abuser as a "stranger" or "unknown abuser" and yet the full context described suggested that the abuser was a friend of a family member. In these conflicting situations, the coders discussed the examples in light of the full transcript. In general, if the situation was one where the participant described the abuser was "known" to the family, but they still categorized the abuser as "unknown" — the abuse was categorized as "unknown" based on the participant's categorization.

• Physical abuse included examples of acts such as hitting, punching, or pushing someone.
• Examples categorized as emotional abuse included situations where participants spoke of intimidation and threatening behaviors and "feeling hurt" by name calling.
• Isolation was coded separately from emotional abuse and referred to examples of situations involving being kept separate from others and being ignored. However, when individuals referred to being "emotionally hurt" by their isolation, the event was categorized as emotional abuse. While isolation could be viewed as a denial of a person's right to socialize, it was kept as a separate category and used to classify examples of being kept separate from others and not being allowed to go out in public and do things.
• Examples categorized as sexual abuse included inappropriate touching or fondling, and unwanted sexual intercourse.
• Financial abuse was used to classify examples given involving use of an individual's funds. It also was used to code situations where people described not receiving monetary raises in their job.
• Examples categorized as neglect included situations where the individual described being abandoned by others. Certain situations where people spoke about being ignored were also categorized as neglect — but only if the participant did not speak about the "emotional hurt" of this type of behavior.
• Denial of rights was used to classify examples involving loss of the right to privacy, the right to bear children, or the right to have a girlfriend or boyfriend.
• In several of the focus groups, participants specifically used the term "discrimination" as a type of abuse and gave examples of being denied the right to a job, an education or a driver's license.

Focus group data are primarily qualitative data — not quantitative data where the respondent selects from pre-established response options. As Krueger suggests (1988), the open-ended questions allow participants to choose the manner in which they respond and the group interaction allows people to change their opinions throughout the discussion. The researcher primarily seeks to identify evidence or themes that are repeated and common to several participants or more. At the same time, the range and diversity of opinions offered in a focus group is vital: hence opinions offered only once are enlightening in demonstrating the range of ideas, but they are not used to form the crux of the findings.

While focus group data are considered qualitative data, software systems have been developed (such as Atlas.ti) that allow the researcher to code themes and run simple frequencies for the data; however, more complex quantitative data analysis is not appropriate.

The data are reported here as frequency distributions of themes discussed in each focus group. For example, if a participant spoke about physical abuse three times in a focus group — even if the reference was to the same situation of physical abuse — each time it came up it was coded as "PHYSICAL." The frequencies reported below, therefore, represent the number of times the code "PHYSICAL" was applied to comments in each focus group as well as totals for all six focus groups. The data presented below, therefore, include multiple references by the same person to a single event.

Sixty (60) people participated in the six focus groups. Groups ranged in size from 6 participants to 23 participants. This included 15 men and 45 women as shown in Table 1. In the focus groups, the ratio of male to female participants ranged from .50 to .80, with the exception of FG3 which was all female and very large. This group therefore skewed the overall gender distribution and resulted in an overall male/female ratio of .33. Several explanations for more female participation are possible: (1) the issue may have resonated as significant for more women and therefore women may have been more likely to agree to participate in the groups or (2) women may simply have been more willing to discuss the issue than men, even though the issue was of significance to both groups. One male participant suggested that a negative stigma about being the subject of abuse may be more of a concern for men than women. He thought it is easier for women to report abuse than men because "… people tend to be more sympathetic towards a woman … that's what they're used to. But if a man's being abused, people tend to say, 'Well, buck up.'… It's kind of like men don't cry, so therefore, there's disbelief. Even though that could be, so I just don't think it goes as reported as much because I don't think … (a) men may not report it as much and (b) because people just might not believe them." (FG2)

No additional demographics were collected on participants for several reasons. First, we thought it would be burdensome to ask participants to complete both a survey and participate in 90-minute focus group. Most had prearranged transportation and time was limited. In addition because of the nature of the discussion, we felt it was better not to collect data on participants that might allow anyone to identify individuals. This was in keeping with concerns expressed by our Institutional Review Board.

Following several focus groups, individuals asked to speak separately with the moderator after the group session. Examples offered in these individual discussions are marked as INT and were added to the data for that focus group.

Table 1. Focus Group Demographics

	FOCUS GROUP
	FG1	FG2	FG3	FG4	FG5	FG6	TOTAL
Male	4	3	0	3	2	3	15
Female	5	4	23	4	4	5	45
TOTAL	9	7	23	7	6	8	60
Ratio Male/Female	.80	.75	0	.75	.50	.60	.33

Findings

Defining Abuse

Table 2 below provides a summary of the different types of abuse that were mentioned in each focus group, including what participants first mentioned (FM) as well as what was mentioned after probing (AP). Examples of emotional abuse were most frequently (N=43) described in the focus groups, followed by physical abuse (N=30), sexual abuse (N=27), neglect (N=18), and financial abuse (N=17). In addition to these more commonly identified types of abuse, examples were also given that one might not commonly think of when considering abuse. These included examples described by participants as denial of their rights (N=21), including abuse of their right to privacy and their right to bear children, have a girlfriend or marry, as well as situations described as discrimination, and isolation (N=8). Finally several examples of abuse were coded as "other." These included references to self-abuse and situations of abuse directed at people who work for someone with a disability. The latter is similar to what has been defined in the literature as "horizontal abuse" where people become targets of abuse by virtue of association (Sims and Napholz, 1996).

In every focus group, participants provided examples of situations involving physical abuse without the need for additional probes. Participants in five of the six focus groups gave examples of emotional abuse and sexual abuse without the use of follow-up probes. Financial abuse was only described in four of the six focus groups without the use of an additional probe. After asking a follow-up question, examples of financial abuse were provided in only one additional focus group. Even with probing, there was no discussion of situations of financial abuse in one focus group. Examples of situations involving neglect were described in four out of the six focus groups. Follow-up probing did not evoke additional examples of neglect in the other focus groups.

Different ways in which a person's rights are denied were provided as examples of abuse in five of the six focus groups. Situations that were described as abuse using the term "discrimination" occurred in two focus groups. Situations described specifically as "isolation" were given in two focus groups. The examples of isolation involved situations where people are kept away from friends, kept in the house, and prohibited from going to public places.

Table 2. "Examples of Things People Do to People with Disabilities that Are Wrong or Bad" — Types of Abuse Participants Spoke About

	Focus group
	FG 1	FG 2	FG 3	FG 4	FG 5	FG 6	Totals
Emotional (FM) Note 1	4	21	0	8	6	4	43
Emotional (AP)	-	-	1	-	-	-	-
Physical (FM)	8	9	2	3	2	6	30
Sexual (FM) Note 2	9	0	4	1	4	7	25
Sexual (AP)	-	2	-	-	-	-	2
Financial (FM)	4	1	0	7	0	2	14
Financial (AP)	-	-	0	-	3	-	3
Neglect (FM)	0	8	0	1	2	7	18
Neglect (AP)	0	-	0	-	-	-	0
Denial of Rights: Total	2	3	0	1	6	9	21
Denial of Rights: Privacy, right to bear children, have a girlfriend, marry	2	3	0	1	2	1	9
Denial of Rights: Discrimination	0	0	0	0	4	8	12
Isolation	3	5	0	0	0	0	8
Other	4	4	0	0	0	1	9

Note 1: FM refers to examples and comments first mentioned without use of probes and AP refers to examples and comments given after probing. Return to Table

Note 2:Includes attempted intercourse Return to Table

Likely Perpetrators of Abuse

From the examples of abuse given by participants, it was possible to learn something about perpetrators of abuse as well. The initial question asking participants to give examples of abuse was followed by another question asking participants to describe the kinds of people who are likely to be abusive towards people with disabilities. As shown in Table 3 below, 123 examples were given involving known perpetrators. The latter included known intimate abusers (N=46) and known acquaintances (N=77). Known intimate abusers included immediate family members (N=28), foster family members (N=5) or intimate partners (N=13). Known acquaintances included staff in institutions and community caregivers including home health aids. In terms of caregivers, more examples were given involving staff in institutional settings (N=23) than caregivers in community settings (N=6). Other known acquaintances included co-workers and bosses (N=14), other people with disabilities (N=12), friends, friends of the family and acquaintances (N=8), fellow students (N=6), teachers (N=1), health professionals such as doctors and nurses (N=4) and police or security guards (N=3). Only 33 examples were given involving unknown perpetrators (e.g., strangers on the street).

Table 3. Perpetrators: Are They Known or Unknown?
(Part One: Known Perpetrators: Intimates)

	Focus group
	FG 1	FG 2	FG 3	FG 4	FG 5	FG 6	Totals
Identified Likely Perpetrators
TOTAL KNOWN INTIMATES	10	13	0	4	10	9	46
Immediate Family Note 1	5	10	0	1	7	5	28
Foster Family	2	1	0	0	1	1	5
Partner	3	2	0	3	2	3	13

Note 1: Immediate and foster family categories include mother, father, siblings, boyfriends of a parent. Return to Table

Table 4. Perpetrators: Are They Known or Unknown?
(Part Two: Known Perpetrators: Acquaintances)

	Focus group
	FG 1	FG 2	FG 3	FG 4	FG 5	FG 6	Totals
Identified Likely Perpetrators
TOTAL KNOWN ACQUAINTANCES	20	16	4	15	7	15	77
Institutional Staff	8	9	3	0	0	3	23
Caregivers in Community	0	1	0	4	0	1	6
Coworkers/Boss	2	0	0	6	1	5	14
Other PWD	5	3	1	3	0	0	12
Friends/Acquaintances	0	1	0	1	1	2	5
Friend of the family	3	0	0	0	0	0	3
Fellow students	2	0	0	1	3	0	6
Teachers	0	0	0	0	1	0	1
Health professionals	0	1	0	0	0	3	4
Police/security guard	0	1	0	0	1	1	3

Table 5. Perpetrators: Are They Known or Unknown?
(Part Three: Summary Ratios of Known vs. Unknown)

	Focus group
	FG 1	FG 2	FG 3	FG 4	FG 5	FG 6	Totals
Identified Likely Perpetrators
TOTAL KNOWN (INTIMATES AND ACQUAINTANCES)	30	29	4	19	17	24	123
TOTAL UNKNOWN	12	4	2	1	7	7	33
Ratio of KNOWN/UNKNOWN	2.5	7.2	2.0	19.0	2.4	3.4	3.8
% of KNOWN/TOTAL	71	88	66	95	71	77	79

In one focus group, participants suggested that abusers could be anyone.

Female: "It can be somebody in your own family."

Male: "It could be professional staff."

Female: "It could be somebody on the street." (FG1)

Some participants had difficulty differentiating between known and unknown perpetrators. As one woman suggested, "The one person, he was a stranger to me, 'cuz I didn't know him that good. The one person that sexually abused me was my mother's boyfriend. I didn't know him that good. So, to me he's a stranger." (FG1)

Discussion

While abuse directed against people with disabilities generally remains hidden from view (it was listed by Project Censored as one of the 25 most underreported stories of 2000) people with disabilities easily identify situations of abuse in the focus groups. They defined abuse much the same way people without disabilities define abuse using examples of physical, emotional, sexual and financial abuse, as well as situations of neglect. They also expanded the definition of abuse to include less commonly considered situations involving limitations of personal rights.

In most situations of abuse provided by participants, the perpetrators were generally known to the person and, in most cases, involved family members. It is important to note that these individuals often serve as major caregivers for persons with disabilities, shouldering significant amounts of responsibility for providing care. While this situation can be understood in terms of the stress this often produces, it should not, however, serve as an excuse for abuse. Participants expressed frustration about abusive caregivers who are excused because of the stress they experience as caregivers. As one pointed out: "'Oh she's stressed out,' like that seems to be more of an excuse for the person who's an abuser because they're care giving someone with a disability, so they must be tired or they're overworked and so that abuse goes on longer because people kind of turn their heads and make excuses when if it were in reverse they probably wouldn't. They would probably be a little more involved." (FG2)

People with disabilities experience marginalization. While people with disabilities are at risk of emotional, physical, sexual and financial abuse, they are likely to be ignored by family, disability and violence-related support systems when they do report abuse. Their marginalization may create a self-destructive sense of anomie or detachment from society and its norms. As one participant lamented, people find it "easier to ignore, neglect somebody who's blind or can't move, or you know, it's easy because people do not listen, you know." (FG6)

An important first step toward addressing this sense of powerlessness and marginalization of people with disabilities is to provide for their voice in the research process itself. In other words, the definitions of abuse should come from those who experience it, not just caregivers and providers. We need to examine the issue of abuse from the perspective of those who experience it. Their descriptions and comments, as shown here broaden traditional definitions, and give a greater reality to otherwise sterile categories. Their insights suggest areas to consider for improving the quality of life of those impacted by abuse. In the past, the issue has been largely studied using only the limited statistics available and data gathered from family members, staff and other caregivers. Our findings suggest that it is important to explore critical issues such as abuse with affected populations themselves.

The findings have implications for research. This research was unique in that every effort was made to include people with disabilities in the research design and data collection process — including involving them in the development of definitions, design of data collection instruments as well as the data collection process. Following data collection, the researchers conducted a follow-up meeting with a group of self-advocacy members to share the findings and obtained their input in terms of next steps. Suggestions included scheduling follow-up meetings with police officers to allow members to discuss their views on abuse. The findings demonstrate the value of using research methods that facilitate the participation of people with disabilities as much as possible. Their contributions served to improve the process overall.

The findings have implications for practice as well. One important way to begin to address concerns about marginalization is to empower people through self-advocacy training — increasing one's understanding of personal rights, of what constitutes abuse — and what is not abuse, as well as the importance of reporting and the process for securing justice. In addition, the findings suggest the need for similar training for families and other caregivers, criminal justice personnel, educators, and employers.

Families, people with disabilities, provider organizations and advocates must work together to enact effective education programs. This educational effort must also target the general public. A wide-reaching, coordinated effort is needed to impact sources of abuse in the community and persons experiencing abuse who remain hidden from view and who are not part of the self-advocacy effort.

Self advocacy projects can collaborate to develop and enact broad training programs with the support of Local Chapters of The ARC and other provider and advocacy organizations in the disability community, including Protection & Advocacy, the Developmental Disabilities Council, university resources and other specific disability organizations. These collaborations are critical. They can encourage discussion among consumers, families, and the larger community. This dialogue is an important first step in the development of training programs and public relations efforts. For example, in New Jersey the Department of the Public Advocate hosts a 'Partners in Justice' committee which includes consumers, providers, and representatives from the criminal justice system. They meet and discuss these issues, share resources, arrange trainings and hold conferences.

Funding for collaborative ventures and training usually falls within the funding criteria of the state's Developmental Disabilities Council. Interested organizations can then provide in-kind support such as meeting space, laptops, projectors, copies, refreshments and staff hours for registration and coordination. Training curricula for self advocates and others have been developed and are easily accessible using the Victims of Crime with Disabilities Resource Guide from the University of Wyoming available at http://wind.uwyo.edu/resourceguide/. The website has a database of training resources and other materials with additional links. Temple University's Institute on Disabilities has also developed curricula for training self-advocates as well as families, caregivers and criminal justice personnel. Their resources are available at http://disabilities.temple.edu/index.htm.

It is important to point out the limitations of this research. The data provided here represent only a summary of the number of times participants used different examples of abuse. The data therefore may include multiple references by the same person to a single event. We made no attempt to tally the actual number of abusive situations discussed and the data are not meant to represent an estimate of prevalence of abuse.

We did not collect demographic data on the participants in the focus groups other than gender based solely on observations. Because our goal was to explore the various definitions of abuse rather than document specific situations of abuse, we felt it was important not to collect personal identifiers on the participants. We wanted to encourage an open discussion. The lack of demographics, however, is a limitation that prevented us from knowing anything about the nature of individual disabilities represented in the sample of participants, or other factors (such as age or living situation) that might be important for data interpretation.

While our research findings reflect stories told from the vantage point of persons with disabilities, it is important to remember that caregivers may also have their own perspective on these stories which is not represented here. We recognize that these situations can be very complex but the purpose of our research was simply to provide the perspective of persons with disabilities since their views are generally not represented in research findings.

The sample of participants in the focus groups was a non-random sample. Participants were people who were participating in pre-existing self-advocacy groups within the community. As such not all people with disabilities in New Jersey had an equal chance of participating in the groups and consequently, the findings cannot be generalized to the population of people with disabilities as a whole. Those who participated were a subset of the population with disabilities. They were people able to attend meetings in the community and willing to participate in a group discussion of this type. These characteristics may make them different from other people who experience greater limitations in their ability to participate in groups. The experiences of people who did not participate may be very different from that of our participants. It is possible that they might have offered different examples of abuse or had no experience with situations of abuse. It is also possible that they might actually experience more abuse than our participants but — because of limited mobility, communication barriers or other factors — they were unable to share their views. If the latter case is the situation, then the findings expressed here may in fact represent only the "tip of the iceberg."