A Brief Report: Interpersonal Violence Exposure and Violence Myth Acceptance in the Ohio Deaf Community

Introduction

Interpersonal violence is a leading cause of death and a prevalent public health issue in the United States, affecting millions of individuals each year (Sumner et al., 2015). Those with disabilities are more greatly impacted, with higher rates of exposure to interpersonal violence and neglect than their nondisabled peers (Nosek, Foley, Hughes, & Howland, 2001). One unique subpopulation of individuals with disabilities, the Deaf¹ community, is two to three times more likely to experience physical violence, sexual violence, bullying, and crime than their non-Deaf, non-disabled peers (Anderson & Leigh, 2011; Anderson, Leigh, & Samar, 2011; Barrow, 2007; Francavillo, 2009; Obinna, Krueger, Osterbaan, Sadusky, & DeVore, 2006; Pollard, Sutter, & Cerulli, 2014; Weiner & Miller, 2006).

One factor that many contribute to Deaf people’s increased exposure to interpersonal violence is a limited understanding of healthy relationship dynamics and nonviolent sexual relations (Anderson & Kobek Pezzarossi, 2012; Elliott Smith & Pick, 2015; Francavillo, 2009; Gilbert, Clark, & Anderson, 2012). These commonly observed health literacy gaps are primarily caused by lack of health education available in American Sign Language (ASL), as well as reduced incidental learning throughout Deaf people’s lifespans – e.g., an inability to communicate with hearing parents, hearing healthcare providers, or understand spoken health information on TV/radio/public service announcements (Francavillo, 2009; Pollard & Barnett, 2009; Pollard, Dean, O’Hearn, & Haynes, 2009).

Stemming from such information deprivation, literature suggests that Deaf individuals are more likely than their hearing peers to possess beliefs that align with common rape myths; i.e., “a specific set of attitudes and beliefs that may contribute to ongoing sexual violence by shifting blame for sexual assault from perpetrators to victims” (Iconis, 2008, p. 47). Compared to rates of rape myth acceptance among hearing individuals, a greater proportion of Deaf people believe that people falsely report rape in order to draw attention to themselves (Francavillo, 2009); that sex within a romantic relationship is one’s obligation and sexual coercion perpetrated by one’s partner is not rape (Anderson & Kobek Pezzarossi, 2012); and, that experiences of sexual violence are better classified as miscommunication or bad sex, rather than rape or sexual assault (Elliott Smith & Pick, 2015).

Although some empirical evidence exists to substantiate Deaf people’s health disparities in interpersonal violence exposure and violence myth acceptance, most prior research on these topics was conducted with college student samples using written English survey measures (for instance, Anderson & Kobek Pezzarossi, 2012; Anderson & Leigh, 2011; Elliott Smith & Pick, 2015); this may be a significant methodological limitation given Deaf people’s median fourthgrade reading level (Gallaudet Research Institute, 2003). Even research efforts that have evolved to data collection via ASL surveys (for example, Pollard et al., 2014) have largely been conducted in the Rochester, New York metropolitan area, where high levels of educational attainment fail to mirror the characteristics of the Deaf community at large and the resulting data, therefore, likely underestimate reported health disparities (Barnett et al., 2011).

To address these limitations, the current secondary analysis leveraged data collected via an ASL survey instrument across a statewide population of hearing individuals and grassroots² Deaf individuals in Ohio. We hypothesized that Deaf participants would report higher rates of interpersonal violence exposure than hearing participants, and that Deaf participants would be more likely to endorse common violence myths than hearing participants.

Method

The current secondary analysis was designed and conducted by the first author in partial fulfillment of the requirements for a doctoral degree. This analysis leveraged archival data collected by Deaf World Against Violence Everywhere (DWAVE), a Deaf-led non-profit organization that serves Deaf grassroots consumers across the state of Ohio. Study procedures were reviewed and approved by Argosy University Institutional Review Board.

Study Overview

DWAVE collected data between 2010 and 2012 via online self-administered surveys located on SurveyGizmo.com. The survey was available to Deaf participants for a period of ten weeks (October 2010 through December 2010) and to hearing participants for a period of 12 weeks, approximately one year later (October 2011 to January 2012).

Participants

Inclusion criteria for Deaf participants were self-reported hearing loss, self-identification as a member of the Deaf community, being 18 years or older, and residing in Ohio. Inclusion criteria for hearing participants were self-identification as hearing (i.e., no hearing loss or no affiliation with the Deaf community), being 18 years or older, and residing in Ohio. No additional criteria were applied in order to recruit diverse samples reflecting the overall sociodemographic characteristics of the Ohio population.

Recruitment and Survey Administration

Convenience sampling was used to extract the Deaf and hearing samples from the Ohio population. For both data collection efforts, study advertisements that included the survey link were distributed via email announcements, social media outlets, and DWAVE’s website. Although recruitment efforts were largely concentrated in the Central Ohio area, individuals from across the state were eligible to participate. DWAVE monitored the online survey website on a weekly basis to ensure that the survey link was working and that the survey mechanism continued to collect data.

Measures

Survey instrument for Deaf participants.

Deaf participants were presented with questions from a revised version of the Community Resources and Needs Assessment. The Regional Prevention Network of Southwest Ohio developed the original Community Resources and Needs Assessment in 2009 to survey teenagers’ rates of exposure to sexual and relationship violence, as well as their access to information about these topics. While several questions corresponded, in part, to questions from the Safe Dates Evaluation Questionnaire, the majority of the measure was born out of the combined experience and training of members of the Regional Prevention Network.

The Regional Prevention Network of Southwest Ohio granted DWAVE permission to adapt its original 70-item measure to better meet the linguistic and cultural need of Deaf grassroots adult survey respondents. DWAVE’s revised measure included a total of 52 items that queried participants’ sociodemographics (13 items), exposure to psychological and emotional violence (10 items), exposure to physical violence (7 items), exposure to sexual violence (2 items), attitudes and beliefs about relationship violence (8 items) and sexual violence (8 items), and help-seeking behaviors (4 items). For the purposes of this report, we focus only on rates of interpersonal violence exposure and violence myth acceptance; help-seeking behaviors will be explored in a subsequent report.

DWAVE’s adaptations included revising the measure to target adults rather than teenage respondents and adding Deaf-related sociodemographic questions (e.g., cultural status, accessibility). More significant adaptations were also made to match the communication needs of the target population. For example, rather than collect data on the frequency of violence exposure or total number of incidents of violence, DWAVE simplified the measure to query only the presence or absence of victimization. Additionally, four-item Likert-scale response options used to evaluate participants’ beliefs and attitudes about dating violence were replaced with a twoitem Agree/Disagree response option.

In addition to these simplifications of the written English survey instrument, DWAVE created ASL videos to further improve accessibility for signing Deaf participants. DWAVE engaged two native ASL users (one Deaf adult, and one adult child of Deaf parents) to conduct the translation. Both were trained in the field of interpersonal violence and collaborated with DWAVE agency staff and work group members to clarify complex concepts and improve translation accuracy. The final translation was performed in ASL, filmed, edited to include English subtitles and voiceover, and embedded into the online survey instrument.

Survey instrument for hearing participants.

Hearing participants were also presented with questions from DWAVE’s revised survey; however, Deaf-related sociodemographic questions were removed. Otherwise, all survey components remained intact across Deaf and hearing data collection procedures. Regardless of hearing status, the survey required approximately one hour to complete.

Statistical Analyses

Data were entered and analyzed in the Statistical Package for the Social Sciences (SPSS) software program, Student Version 20.0. For all analyses, statistical significance was set at p ≤ .05.

The first hypothesis was that Deaf participants would report higher rates of interpersonal violence exposure than hearing participants. For the purposes of the current analyses, “violence exposure” was measured on a dichotomous scale (1 = Yes and 2 = No), and was defined as any reported experience of psychological, emotional, physical, or sexual violence. To test this hypothesis, Pearson’s chi-square tests of independence were conducted, with hearing status (Deaf or hearing) as the independent variable and specific examples of violence exposure as each of the dependent variables, resulting in a series of 2 × 2 contingency tables. As such, the Yates Continuity Correction was applied to account for the fact that the Pearson’s chi-square test is biased upwards for 2 × 2 contingency tables. Additionally, for any chi-square analyses that violated the assumption that each cell has expected frequencies of five or more, Fisher’s Exact Test was applied.

The second hypothesis was that Deaf participants would be more likely to endorse common violence myths than hearing participants. “Violence myth acceptance” was measured on a dichotomous scale for beliefs about relationship violence (1 = Agree and 2 = Disagree) and beliefs about sexual violence (1 = True and 2 = False). To test this hypothesis, Pearson’s chi-square tests of independence were conducted, with hearing status (Deaf or hearing) as the independent variable and specific examples of violence myth acceptance as the dependent variables. The Yates Continuity Correction and Fisher’s Exact Test were again applied as described above.

Results

Sample Characteristics

One hundred eighty-six individuals participated in the current study, with 75 identifying as Deaf and 111 identifying as hearing. Additional sample characteristics, separated by hearing status, are outlined in Table 1.

Table 1:

Sociodemographic Characteristics of the Study Sample (N = 186)

Sociodemographic Characteristic	% by Hearing Status Deaf (n = 75)	Hearing (n = 111)
Gender
Male	29.3	9.0
Female	70.7	91.0
Age
18 – 34 years	25.3	34.2
35 – 50 years	49.3	36.9
51+ years	25.3	28.8
Race/Ethnicity
White/Caucasian	88.0	85.6
African American	5.3	10.8
Hispanic/Latino/Latina	2.7	0.0
Other	1.3	0.0
Biracial	2.7	2.7
Did Not Answer	0.0	0.9
Relationship Status
Single	30.7	26.1
Married	46.7	39.6
Partnered	14.7	22.5
Divorced	8.0	9.0
Widowed	0.0	2.7
Residence
Central Ohio	85.3	59.5
Other Areas of Ohio	14.7	39.6
Did Not Answer	0.0	1.1
Hearing Loss
No Hearing Loss	-	94.6
Hard of Hearing	-	5.4
Deaf Identity
Deaf	80.0	-
Hard of Hearing	16.0	-
Deaf-Blind	2.7	-
Both Deaf and Hard of Hearing	1.3	-
Preferred Mode of Communication
Sign Language	98.7	-
Oral/Speechreading	1.3	-
Written	0.0	-
Prior Participation in Workshops on
Relationship and Sexual Violence (Yes)	46.7	73.9

Sociodemographic characteristics were relatively comparable across levels of hearing status, with a few exceptions. The Deaf sample had a larger proportion of male participants than the hearing sample (29.3% versus 9.0%, respectively); this sampling discrepancy is explored in the Discussion section. Although it appeared that Deaf males (n = 20) in the sample were older than hearing males (n = 12) in the sample – 30.0% of Deaf males age 51 and up versus 8.3% of hearing males age 51 and up – subsample sizes were too small to calculate any reliable differences between these subgroups.

The Deaf sample also had a larger proportion of participants from within central Ohio compared to the hearing sample (85.3% versus 59.5%, respectively). Additionally, the Deaf sample reported lower exposure to information about relationship and sexual violence via workshops compared to hearing participants (46.7% versus 73.9%, respectively).

Interpersonal Violence Exposure

Rates of exposure to specific examples of psychological, emotional, physical, and sexual violence are outlined below in Table 2. Chi-square results indicated no significant differences based on hearing status when comparing each example of violence exposure.

Table 2:

Rates of Interpersonal Violence Exposure, Separated by Hearing Status(N = 186)

	% by Hearing Status		p-value
	Deaf (n = 75)	Hearing (n = 111)
Damage to Property	54.7	51.4	.679
Hurt Feelings	82.7	83.8	1.000
Public Insult	62.7	66.7	.507
Threatened to Leave	37.3	33.3	.702
Simulated Hit	61.3	50.5	.102
Made Jealous	50.7	50.5	1.000
Abusive Behavior	50.7	53.2	.805
Threatened with Weapon	17.3	18.9	.935
Criticized Appearance	65.3	54.1	.167
Made Uncomfortable	45.3	45.9	1.000
Pushed, Grabbed, or Shoved	56.0	61.3	.810
Bitten	6.7	7.2	1.000
Physically Stopped Departure	36.0	38.7	1.000
Forced Sex	29.3	27.9	.691
Forced Other Sexual Acts	17.3	26.1	.306
Choked	14.7	21.6	.458
Physically Assaulted	37.3	43.2	.783
Beat Up	17.3	20.5	.908
Hurt With Weapon	6.7	8.1	1.000

Violence Myth Acceptance

Rates of endorsement of specific attitudes and beliefs that support relationship and sexual violence are outlined below in Table 3.

Table 3:

Rates of Violence Myth Endorsement, Separated by Hearing Status (N = 186)

	% by Hearing Status		p-value
	Deaf (n = 75)	Hearing (n = 111)
Hit Partner When Angry	5.3	0.0	.020
Deserve to be Hit	8.0	0.0	.003
Hit Back if Partner Hits First	18.7	9.0z	.059
Equal Power	82.7	98.2	.028
One Person Makes All Decisions	5.3	0.9	.067
Send Naked Photos	8.0	9.9	1.000
Expect to Know Partner’s Whereabouts	38.7	11.7	< .001
Sexting	13.3	14.4	1.000
Rape Can Happen to Anyone	82.7	98.2	.030
Rape Due to Appearance & Behavior	54.7	30.6	< .001
Rape Due to Perceived LGBTQIA	40.0	49.5	.668
Status
Most Victims Know Their Attacker	61.3	89.2	< .001
Rape Due to Alcohol or Drugs	16.0	0.0	< .001
Okay to Say “No” to Previous Sexual	85.3	100.0	.138
Partner
Person Paying for Date Deserves Sex	2.7	0.0	.143
Rape is Never the Victim’s Fault	65.3	97.3	< .001

Chi-square results indicated that Deaf respondents had significantly higher rates of violence myth acceptance in nine of the 16 possible areas. Specifically, compared to hearing participants, a higher proportion of Deaf participants reported that they believed that it is okay to hit one’s partner when angry, that sometimes people deserve to be hit, that they expect their partner to tell them where they are at all times, that rape can be caused by the victim’s appearance or behavior, and that rape can be caused by the victim’s use of alcohol or drugs. Additionally, Deaf participants were less likely than hearing participants to support the concept of equal power in relationships, to believe that rape can happen to anyone, that most victims know their attacker, and that rape is never the victim’s fault.

Discussion

The present secondary analysis compared rates of interpersonal violence exposure and violence myth acceptance between Deaf and hearing samples extracted from the Ohio general population. Compared to previous research efforts, the current study is relatively unique in its use of an ASL-accessible survey administered outside of the Rochester, New York, metropolitan area.

Interpersonal Violence Exposure

The first hypothesis was that Deaf participants would report higher rates of interpersonal violence exposure than hearing participants. Contrary to expectations and prior literature (Anderson & Leigh, 2011; Anderson et al., 2011; Barrow, 2007; Francavillo, 2009; Obinna et al., 2006; Pollard et al., 2014; Weiner & Miller, 2006), this hypothesis was not supported. Rather, rates of violence exposure were largely similar across Deaf and hearing samples.

One potential reason for the discrepancy between our results and prior literature could be the ASL accessibility of the current survey instrument. If this indeed is the case, it would suggest that previous data collected via written English measures could have overinflated rates of violence exposure due to participants’ lack of comprehension of survey items. Yet, recent data collected using an ASL public health survey also identified Deaf people’s disparities in intimate partner violence exposure (Pollard et al., 2014), likely disproving this theory.

A more likely reason for this discrepancy could be the gender difference observed between our hearing and Deaf sample, a frequent result of convenience sampling. Ninety-one percent of the hearing sample identified as female, compared to only 70% of the Deaf sample. Given that women are at higher risk than men for several forms of interpersonal violence, such as kidnapping, physical assault by an intimate partner, rape, sexual assault, and stalking (Iverson et al., 2013), and given the significantly smaller proportion of women in the Deaf sample, rates of violence exposure reported by the Deaf sample could have been artificially underestimated due to this gender discrepancy.

Another possible reason underlying the departure from previous literature could also be variable influence of self-selection bias into the Deaf and hearing samples. Hearing individuals who chose to participate in the study were likely drawn by the relevance of the study topic to their lives and, therefore, more likely to be survivors of interpersonal violence. Deaf individuals who chose to participate, however, may have been more drawn to the accessibility of the measure in ASL than the survey topic. If these potential differences in self-selection did occur, it could have obscured any true disparities in interpersonal violence exposure between the groups.

Violence Myth Acceptance

The second hypothesis was that Deaf participants would be more likely to endorse common violence myths than hearing participants. Current results support this hypothesis, with Deaf participants more likely to endorse a number of myths about relationship and sexual violence compared to hearing participants. In other words, although rates of violence exposure were similar between our Deaf and hearing samples, Deaf participants were more likely to blame themselves and other Deaf victims for their experiences of victimization, rather than shift the blame to the perpetrator of those violence.

This finding can be explained, in part, by the low levels of exposure to domestic violence and sexual violence workshops reported by Deaf participants in our study. Nearly three-fourths of the hearing sample had received exposure to this critical health information via workshops, compared to less than half of the Deaf sample. Although we did not directly investigate this hypothesis, reduced exposure to information about healthy relationships and healthy sexual relations has been previously linked to Deaf people’s low health literacy in these areas (Francavillo, 2009; Pollard & Barnett, 2009; Pollard et al., 2009).

It is also possible that the observed differences in violence myth acceptance were, in part, an artifact of our current dataset specifically related to the gender difference described above. Research suggests that men are more likely to endorse common rape myths than women (Hayes, Lorenz, & Bell, 2013; Johnson, Kuck, & Schander, 1997). It is, therefore, possible that some of our observed disparities in violence myth acceptance are due to the greater male make-up of the Deaf sample.

Future Empirical and Clinical Directions

Compared to previous literature on violence in the Deaf community, the present study incorporated additional methodological strengths, including grassroots, community-driven survey development and data collection efforts, use of an ASL-accessible survey, collection of a relatively large Deaf sample, and avoidance of a highly-educated, university-adjacent Deaf sample.

Despite these improvements, future empirical efforts in this area should apply rigorous sampling techniques that are likely to result in Deaf and hearing samples that are more closely balanced on key demographic variables that could confound results, such as gender, age, race/ethnicity, and educational attainment. Whenever possible, stratified sampling procedures should be applied to achieve greater sociodemographic representativeness of the general population, as neither sample in the current study represented the gender make-up of Ohio residents (51% female, according to 2010 and 2016 Census data). Although such rigor can be challenging to apply when recruiting from an especially small population like the Deaf community, it may allow for greater confidence in the interpretation and application of our field’s findings.

With regard to clinical implications, 54% of the Deaf sample reported that “some people get raped because of their appearance and behavior.” Forty percent reported that “some people get raped because they are, or seem to be homosexual.” Thirty-five percent do not agree that that “rape is never the victim’s fault.” And 15% do not believe that “it is okay to say ‘no’ to having sex even if you have had sex with that person before.” These numbers are alarming, regardless of any comparison to hearing participants or our sample’s gender composition. Combined with the low level of access to domestic violence and sexual violence workshops reported by the Deaf sample, current results call for increased psychoeducation efforts that specifically target members of the Deaf community. Similar to the approaches used in this study, we specifically recommend the application of community-engaged methodologies through which Deaf survivors of interpersonal violence guide the development and implementation of psychoeducational efforts for their own peers.