Prevalence of Sexually Abusive Experiences in Childhood and Adolescence among a Community Sample of Latinas: A Descriptive Study

Abstract

This study examined the prevalence, severity, and circumstances of self-reported sexually coercive and abusive experiences in childhood and adolescence in a community sample of Latina women (N = 204) ages 18 to 34 years. Structured phone interviews were conducted in either English or Spanish. Results indicated that 35% of the women reported experiencing some form of sexual abuse; 31% of the reported perpetrators were family members, and 52% were boyfriends, friends or acquaintances. Of those who reported an abusive experience, 44% said they had not disclosed the abuse to anyone; for those who did disclose, 74% said they felt supported. Findings highlight the importance of including peer and dating-partner abuse in the assessment of CSA, and the necessity for outreach to Latina women who may have never disclosed their experiences of abuse.

The purpose of this study was to expand the knowledge base pertaining to Latina women’s experiences of childhood and adolescent sexual abuse (CSA) by examining the prevalence, circumstances, and severity of both disclosed and undisclosed sexually coercive and abusive experiences in childhood and adolescence in a community sample of Latina women of predominantly Mexican descent. Little research on circumstances and severity of CSA among Latina women exists. Thus, prevalence rates of CSA among Latinas are not well established. Furthermore, researchers have highlighted the need to conduct cultural and ethnic specific research on CSA (Arellano, Kuhn, & Chavez, 1997; Arroyo, Simpson, & Aragon, 1997). This is particularly important for Latinos/as since they are the fastest growing ethnic group in the United States (U.S. Census Bureau, 2007). Additionally, examining the rates and circumstances of a community sample of Latina women facilitates the ability to design and implement mental health and social services that serve this community.

Prevalence of CSA

Several studies of CSA have included Latina women or children and compared their rates of abuse to those of other ethnic groups. Some studies have found ethnic differences in prevalence of CSA (Arellano et al., 1997; Ullman & Filipas, 2005) whereas others have not (Arroyo et al., 1997; Mennen, 1994). For example, in Arroyo and colleagues’ study of Latina and White college women, 31.2% of the total sample experienced some form of CSA: 27.1% of Latina women, and 33.1% of White women. These rates are comparable to other studies with community samples of Latina women of Mexican descent (i.e., 33% in Romero, Wyatt, Loeb, Carmona, & Solis, 1999) and Mexican-American adolescent females (i.e., 20% in Arellano et al.). Most estimates of the prevalence of CSA in the general female population range from 15% to 33% (Briere & Runtz, 1988; Finkelhor, Hotaling, Lewis, & Smith, 1990; Saunders, Villeponteaux, Lipovsky, & Kilpatrick, 1992; Wyatt, 1985) depending on how CSA is measured and defined.

Definitions and Measurement of CSA

Childhood sexual abuse has been defined and measured in a variety of ways in psychological research. CSA has been defined primarily by age rather than by the nature of the sexual activity, with the cut-off age typically ranging from 12 to 20 years old (Miller, 1999). To further complicate matters, the manner in which sexual abuse has been measured varies across studies. Most commonly, women are dichotomized into two groups: (a) those who report having been sexually abused and (b) those who have not (Roosa, Reyes, Reinholtz, & Angelini, 1998). Other methods have taken into account the severity of the experience (i.e., distinguishing between varying degrees of unwanted sexual contact) and have included non-contact experiences such as being forcibly exposed to nudity or pornographic materials. In addition, the age of the perpetrator has been a factor in defining childhood sexual abuse. For example, some studies have not included forced or unwanted sexual events with similar-aged peers in their definitions of childhood sexual abuse (see Roosa et al. for a review).

As a result of these varying definitions and methods of measurement, estimates of the prevalence of CSA also vary. This study defined childhood sexual abuse as any unwanted sexual experiences occurring before age 18, regardless of the relationship to the perpetrator. It also utilized a multi-level measure of sexual abuse that has been shown to provide better information to researchers about occurrence and severity of sexual abuse (Roosa et al., 1998).

Circumstances of Abuse

Little research since the late 1990s has examined the detailed circumstances and severity of CSA among Latina women. In one of the few studies specifically examining the circumstances of CSA among Latina women, Romero et al. (1999) found the following: (a) 35% of the women who reported being abused experienced more than one incident; (b) the mean age of the victim at the time of abuse was 11; (c) 48% were abused by individuals within their family; (d) most (60%) did not disclose the abuse to anyone; and (e) less acculturated women were less likely to disclose the abuse. The examination of the unique circumstances of CSA among Latina women such as acculturation, relationship to perpetrator, and women’s roles in the family may provide information central to understanding their reactions to and disclosure of CSA.

It is consistently reported in the literature that compared to other ethnic groups, Latina women and girls are more likely to be abused by male family members or relatives such as uncles, cousins, brothers, grandfathers, fathers, and stepfathers (Huston, Parra, Prihoda, & Foulds, 1995; Katerndahl, Burge, Kellogg, & Parra, 2005; Romero et al., 1999; Shaw, Lewis, Loeb, Rosado, & Rodriguez, 2001; Ullman & Filipas, 2005). In addition, Romero and colleagues found that four of the women in their study were forced by their families to marry the perpetrators of their abuse. Latino family values may contribute to the impact of CSA either positively or negatively (Comas-Diaz, 1995). For example, members of Latino families, including the extended family, are expected to be a source of emotional support for one another. This familial support and interpersonal closeness (familismo) may be very helpful for CSA survivors (Hinson, Koverola, & Morahan, 2002). On the other hand, a child who is abused by someone deserving of respect (e.g., a father or stepfather) may be reluctant to disclose the abuse out of respect for the abuser and fear of the consequences within the family (Comas-Diaz). In this case, the cultural value of respect (respeto) may increase the likelihood that the abuse continues due to the victim's fear of exposing the respected family member (Comas-Diaz; Hinson et al.). Other cultural constraints such as placing the family's needs above one’s own needs or wishes (a facet of the cultural value familismo) is another possible barrier to disclosure, especially if the perpetrator is another family member. In addition, the tendency for Latinos/as to keep their problems within the family can function as a barrier to disclosure and seeking outside social support (Hinson et al.).

Rates of disclosure among Latino/a victims of sexual abuse vary in the literature. In a study of multi-ethnic college students, Ullman and Filipas (2005) found that 70.6% of their sample had disclosed their experiences of abuse. However, the study did not report disclosure rates by ethnicity, so whether or not rates for the Latina women were lower is unknown. In the Arroyo et al. (1997) study of Latina and White college students, 23.5% of the Latina women had never disclosed, and 28.3% of the White women had never disclosed their abuse. The authors noted that this ethnic difference was non-significant. In addition, they reported no significant ethnic differences in timing of or reaction to the disclosure.

One the other hand, in a study of 159 sexually abused African American and Latina girls ages 6 to 18, Shaw and colleagues (2001) found that the Latina girls in the sample waited longer to disclose the abuse. Shaw et al. also found that for the Latina girls, the perpetrators were significantly older and more likely to be fathers or stepfathers. In Romero and colleagues’ (1999) study of Latina women, 60% of those who had been abused said they had not disclosed their abuse. When asked reasons for non-disclosure, some women reported the following: (a) anticipating negative responses (i.e., not being believed, being blamed or physically injured); (b) not having anyone to tell; (c) not wanting to cause trouble or upset the family; and (d) not wanting anyone to know about the incident because they felt ashamed or embarrassed.

Although the decision to disclose can be complex and the reactions by others may vary, it may be an important step in order to obtain much needed social support and closure. Latino families often have high levels of interpersonal closeness and involvement which can be a strong source of social support or stress for survivors of CSA (Comas-Diaz, 1995). Few studies have asked Latina CSA survivors about the results of their decisions to disclose abuse. However, in the Ullman and Filipas (2005) study, 44.9% of the women who reported experiencing abuse said there was a positive outcome, 38.2% said there was no difference, and 16.8% said the disclosure made things worse.

Because little research on circumstances and severity of CSA among Latina women exists, the purpose of this current study was to examine the prevalence and characteristics of childhood and adolescent sexually coercive and abusive experiences among a community sample of Latinas. Information regarding age at which they first experienced each abusive event, severity of sexual abuse experiences, relationship to perpetrator, disclosure, and whether or not they felt supported if they disclosed their abuse was obtained. Based upon previous research by Romero et al. (1999), it was predicted that there would be high rates of non-disclosure and family perpetrated sexually abusive events in this sample.

Method

Participants

The participants were 204 women, ages 18 to 34 years old, living within San Diego County in California. The average age of participants was 25.04 years (SD = 4.2). The majority of the women identified themselves as either Mexican (61%, n = 124) or Mexican American (33.8%, n = 69). The other 5.5% (n = 11) identified themselves as Central American, Caribbean, or of other Latin American descent. Interviews were conducted in the language preferred by the women; 53% (n = 108) were in Spanish, and 47 % (n = 96) were in English. Generational status in the U.S. was also assessed and had the following breakdown: (a) 57% (n = 115) were born in Mexico or another Latin American country (i.e., first generation); (b) 34% (n = 69) were second generation (born in the U.S. with either parent born in another country); (c) 5% (n = 10) were third generation (born in the U.S. and parents born in the U.S., but all grandparents born in another country); and (d) 5 % (n = 10) were fourth generation or beyond (born in the U.S., parents born in the U.S., and at least one grandparent born in the U.S.). Additional demographic information is presented in Table 1.

Table 1.

Demographics of Sample

	N	% of Total Sample	Mean	Standard Deviation
Age	204		25.04	4.2
Ethnic Identity
Mexican	124	61
Mexican American	69	33.8
Central American	4	2
Caribbean	3	1.5
Other	4	2
Language Preference for Interview
Spanish	108	53
English	96	47
Generation Status
1st	115	56.4
2nd	69	33.8
3rd	10	5
4th	4	2
5th	6	3
Job Status
Working full-time	36	18
Working part-time	32	15.7
Unemployed	23	11.5
Full-time homemaker	101	49.5
Student	9	4.4
Disabled	2	1
Other	1	.5
Education
No school	2	1
11th grade or lower	85	42
High school degree or GED	70	34
Some college or trade school	33	16
College degree	14	7
Income per Month For Household
Less than $500	19	9.4
$500-$1000	43	21.2
$1001-$1500	44	22
More than $1500	71	35
Don’t know or supported by others	26	12.8
Missing	1	.5
Sexual Orientation
Sex always with men	190	95
Sex almost always with men	8	4
Sex with men and women equally	2	1
Sex almost always with women	0	0
Sex always with women	0	0
Marital Status
Never married/not living with partner	38	18.6
Not married/living with partner	58	28.4
Married	87	42.6
Separated or Divorced	21	10.3

Procedure

Participants were recruited from six different branches of a program providing nutrition and health education services for low income women and children (i.e., $1500 a month for a one-person household) in the San Diego County area. The sites were chosen because of their predominantly Latina clientele. Handouts containing a brief screening questionnaire and detailed information about the study were available at each of the sites to all women who wished to participate. The intent of the screening questionnaire was to obtain contact information and to screen for age and ethnicity requirements. Age was restricted to 18 to 34 years because of the parent study’s focus on Latina women at highest risk for HIV. All materials were available in English and Spanish. The staff at each site agreed to assist women who had difficulty reading, writing, or understanding the questionnaire. Women who completed the screening questionnaire and met the eligibility criteria were then contacted by phone to further participate in the study.

A total of 284 completed screening questionnaires were collected from the six sites. Of those, 204 women (71.8%) met eligibility criteria and were interviewed. The remaining 80 women were either not eligible or were not able to be contacted by phone for various reasons (i.e., disconnected phone numbers, not being available when called, etc.). Of the eligible women who were able to be contacted, none declined to participate in the study. Participants who completed the interviews received a $20 gift certificate to a discount department store.

Measures

Demographic characteristics were assessed with questions about age, ethnic identity, generation status, job status, level of education, income, sexual orientation, and marital status.

Sexually coercive and abusive experiences in childhood and adolescence were assessed using a modified version of the Sexual Experiences Survey (SES; Koss & Oros, 1982; Roosa, Reyes, Reinholts, & Angelini, 1998). The SES has been shown to be a reliable and valid method of assessing sexual abuse. The scale consists of 12 yes/no questions about the occurrence of contact molestation, coercion, attempted rape, and rape. Roosa and colleagues (1998) modified the original SES by simplifying the language to be understood by a wider range of educational and reading levels (in both English and Spanish) for a multi-ethic, community sample of women in Arizona.

In addition, Roosa et al. (1998) suggested a multi-level categorization system of sexual abuse in order to provide more information about women’s sexual abuse experiences and the processes that contribute to women’s adjustment for each specific type of abuse. This six-level measure of sexual abuse has been previously used by Boyer and Fine (1992). As suggested by Roosa et al. (1998), the women in this study were assigned to the category representing the most severe level of sexual abuse they reported experiencing. Wyatt (1985) also recommended analyzing the most severe incident of abuse when multiple incidents of abuse are reported. The six levels of severity of childhood and adolescent sexual abuse (experiences that occurred before age 18) were as follows: (a) 0 = no sexual abuse, (b) 1 = went farther sexually than wanted to go, (c) 2 = coercion, (d) 3 = contact molestation, (e) 4 = attempted rape, and (e) 5 = rape. Reported abuse experiences from women who were ages 18 and 19 years old at the time of the interview were examined to make sure the coercive and or abusive events had not happened within the past year. The shortest time period between age of abusive event and current age was 2 years. Therefore, childhood and adolescent sexually abusive events could be said to have happened in the past as opposed to very recently (within the past 2 years) for women who just turned 18 years old.

Reports of internal consistency (.74) and test-retest reliability (.93) for the SES have been good (Koss & Gidycz, 1985). Roosa et al. (1998) reported Cronbach’s alpha of .84 using 9 of the 12 items. This study yielded a Cronbach’s alpha of .85 with all 12 items.

In order to obtain additional information about each reported abusive event, additional questions were asked such as age when they first experienced the sexually coercive/abusive event, their relationship to the perpetrator, whether or not they ever disclosed the abuse event and to whom, and whether or not they felt supported in their disclosure of the abuse. If the respondents answered yes to experiencing any of the abuse events, they were then asked at what age they first experienced this event, and whether or not the event happened more than once (i.e., a yes/no item asking “Did this event happen more than once?”). Response choices for the questions about perpetrators and to whom they disclosed the abuse were: “boyfriend,” “friend/acquaintance,” “authority figure,” “family member,” “stranger,” or “other.” If respondents choose “other” they were asked to specify. Responses to the disclosure and support questions were yes/no.

Translation and back translation of the SES and the demographics items were conducted by three native Spanish-speaking, bilingual research assistants. First, one of the assistants translated the English materials into Spanish. Second, another assistant back translated the Spanish materials back into English. Third, the few differences between the Spanish and English versions were resolved by the third assistant, acting as an independent judge. The principal researcher, who was also English/Spanish bilingual, then provided the final review that all translated items retained the original meaning.

Data Analyses

Descriptive statistics and frequencies were examined for the demographic variables and items in the SES. Chi-square tests were used to assess for differences in demographic characteristics, perpetrators of the abuse events, disclosure, and support for disclosure by severity level of the sexually coercive and abusive events.

Results

Demographics

As shown in Table 1, the mean age of this sample was 25.04 (SD = 4.2), the majority were full-time homemakers (49.5%), and 33.7% were working either full or part-time. Only 34% of the sample had earned a high school degree, and 43% completed less than the 12^th grade. The sample was relatively low income, with 52.6% of the households earning less than $1500 a month.

Prevalence of Self-reported CSA

Prevalence and circumstances of sexually coercive and abusive experiences are presented in Table 2. In the total sample, 35% of the women (n = 71) reported some form of sexual abuse experience in childhood or adolescence. The women were categorized into the most severe level of abuse they experienced. The breakdown was as follows: (a) 19% (n = 38) reported rape as the most severe sexual abuse experience; (b) 2% (n = 5) reported attempted rape as the most severe experience; (c) 4% (n = 8) had experienced contact molestation, (d) 4% (n = 8) had experienced coercion, and (e) 6% (n = 12) reported someone having gone farther sexually than they wanted as their most severe sexual abuse experience during childhood or adolescence.

Table 2.

Prevalence and Circumstances of Sexually Coercive and Abusive Events

	N	%	Mean	Standard Deviation
CSA Severity
Went farther sexually	12	6
Coercion	8	4
Contact Molestation	8	4
Attempted Rape	5	2
Rape	38	19
Age first experienced most severe form of CSA			13	4.26
Perpetrator
Boyfriend	19	27
Friend or acquaintance	18	25
Authority figure	1	1.4
Family member	22	31
Stranger	10	14
Abuse happened once	38	54
Abuse happened more than once	33	47
Disclosure
Yes	40	56
No	31	44
Who they told
Boyfriend	4	10
Friend or acquaintance	6	15
Authority figure	0	0
Family member	25	63
Stranger	0	0
Felt supported
Yes	29	74
No	10	26

Circumstances Surrounding Most Severe Experience of Abuse

The average age at which the women reported first experiencing their most severe form of sexual abuse was 13 (SD = 4.26, range = 2 to 17). The perpetrators listed were as follows: (a) a family member (31%, n = 22), (b) a boyfriend (27%, n = 19), (c) a friend or acquaintance (25%, n = 18), (d) a stranger (14%, n = 10), and (e) an authority figure (1.4%, n = 1). A little more than half of the women reported that the most severe abuse only happened once (54%, n = 38), and 47% (n = 33) reported that the abuse occurred on more than one occasion.

Of the women who reported experiencing a sexually coercive or abusive event, 56% (n = 40) said they told someone about the event, and 44% (n = 31) said they did not. When asked to whom they disclosed this information, 63% (n = 25) said a family member, 15% (n = 6) said a friend, and 10% (n = 4) reported a boyfriend. In addition, the women were asked if they received adequate social support in regards to their disclosure of the event. Approximately, three quarters (74%, n = 29) of the women said they felt supported, and 26% (n = 10) said they did not feel supported.

Chi-square analyses were performed to assess differences in circumstances of coercive and abusive events by level of severity (see Table 3). Severity of abuse events significantly differed by perpetrator, χ² (16, N = 71) = 29.05, p = .02; disclosure, χ² (4, N = 71) = 14.89, p = .005; and generation in the U.S, χ² (20, N = 71) = 36.87, p = .01. Due to the seriousness of childhood rape and the significant differences in severity of CSA by perpetrator, a closer examination of the circumstances of rape was warranted. An additional chi-square analysis was conducted to examine whether the perpetrator of the most severe level of sexual abuse experience (i.e., rape) varied as a function of the age of the victim when rape first occurred (e.g., rape by a boyfriend or friend versus rape by a family member). For this analysis age at which rape first happened was dichotomized into rape before age 13 or after age 13. Results indicated age at which rape first occurred differed significantly by perpetrator, χ² (4, N = 38) = 16.57, p = .002 (see Table 4). For women who first experienced rape before age 13, friends/acquaintances (n = 9) were the most frequently reported perpetrators, followed by boyfriends (n = 5), then strangers (n = 4). Among women who first experienced rape after age 13, family members (n = 12) were the most frequently reported perpetrators followed by friends/acquaintances (n = 3) and strangers (n = 2).

Table 3.

Circumstances by Severity Group

Variable	No abuse	Went Farther Sexually	Coercion	Contact Molestation	Attempted Rape	Rape	Total	χ2
Generation in U.S.	(133)	(12)	(8)	(8)	(5)	(38)	(304)	36.87**
1st	(77) 57.9	(7) 58.3	(2) 25	(6) 75	(3) 60	(20) 52.6	(115) 56.4
2nd	(45) 33.8	(5) 41.7	(5) 62.5	(0) 0	(1) 20	(13) 34.2	(69) 33.8
3rd	(6) 4.5	(0) 0	(0) 0	(0) 0	(0) 0	(4) 10.5	(10) 4.9
4th	(3) 2.3	(0) 0	(1) 12.5	(0) 0	(0) 0	(0) 0	(4) 2.0
5th	(2) 1.5	(0) 0	(0) 0	(2) 25	(1) 20	(1) 2.6	(6) 2.9
Perpetrator		(11)	(8)	(8)	(5)	(38)	(70)	29.05*
Boyfriend		(3) 27.3	(8) 100	(1) 12.5	(2) 40	(5) 13.2	(19) 27.1
Friend or acquaintance		(3) 27.3	(0) 0	(3) 37.5	(0) 0	(12) 31.6	(18) 25.7
Authority figure		(0) 0	(0) 0	(0) 0	(0) 0	(1) 2.6	(1) 1.4
Family member		(3) 27.3	(0) 0	(3) 37.5	(2) 40	(14) 36.8	(22) 31.4
Stranger		(2) 18.2	(0) 0	(1) 12.5	(1) 20	(6) 15.8	(10) 14.3
Disclosure		(12)	(8)	(8)	(5)	(38)	(71)	14.89**
Yes		(7) 58.3	(8) 100	(3) 37.5	(1) 20	(12) 31.6	(31) 43.7
No		(5) 41.7	(0) 0	(5) 62.5	(4) 80	(26) 68.4	(40) 56.3
Who they told		(5)	(0)	(5)	(4)	(25)	(39)	7.01
Boyfriend		(0) 0	(0) 0	(0) 0	(0) 0	(4) 16	(4) 10.3
Friend		(1) 20	(0) 0	(1) 20	(0) 0	(4) 16	(6) 15.4
Authority Figure		(0) 0	(0) 0	(0) 0	(0) 0	(3) 12	(3) 7.7
Family		(4) 80	(0) 0	(4) 80	(4) 100	(13) 52	(25) 64.1
Stranger		(0) 0	(0) 0	(0) 0	(0) 0	(1) 4	(1) 2.6
Felt Supported		(5)	(0)	(5)	(4)	(25)	(39)	2.34
No		(1) 20	(0) 0	(0) 0	(1) 25	(8) 32	(10) 25.6
Yes		(4) 80	(0) 0	(5) 100	(3) 75	(17) 68	(29) 74.4

Percentages (and n Values);

^*p < .05;

^**p < .01

Table 4.

Perpetrator by Age at Which Most Severe Abuse First Happened

Variable		Rape <13	Rape =>13	Total	χ2
Perpetrator		(21)	(17)	(38)	16.57**
Boyfriend		(5) 23.8	(0) 0	(5) 13.2
Friend		(9) 42.9	(3) 17.6	(12) 31.6
Authority		(1)	(0)	(1)
Figure		4.8	0	2.6
Family		(2) 9.5	(12) 70.6	(14) 36.8
Stranger		(4) 19	(2) 11.8	(6) 15.8

Percentages (and n Values);

^*p < .05;

^**p < .01

Additional chi-square analyses examined differences in disclosure. Whether or not they felt supported did not vary by whom they told, χ² (4, N = 38) = .65, p = .96. There were also no significant differences between disclosure and generational status or language preference.

Discussion

This study provided information on the prevalence, circumstances, and severity of sexually coercive and abusive experiences in childhood and adolescence within a community sample of Latina women of predominantly Mexican descent. Very little research in the area of childhood and adolescent sexual abuse has focused on issues specific to Latina women. Consistent with the few other studies of childhood and adolescent sexual abuse among Latina samples, this study found high rates of non-disclosure, and family members were the most frequently reported perpetrators of abuse followed by boyfriends and friends or acquaintances. However, when examining rape only, women who first reported rape before age 13 most frequently listed friends as perpetrators, whereas those who first experienced rape after age 13 reported family members as the perpetrators. Findings from this study highlight the importance of assessing for peer and dating-partner related sexually abusive events, assessing for both reported and non-reported experiences of abuse, and seeking social support from family members.

The prevalence of self-reported sexually coercive and abusive events in childhood and adolescence found in this study (35%) were consistent with self-reported CSA rates in a community sample of Latina women in Southern California (Romero et al., 1999) but were slightly higher than CSA rates found with Latina college women (Arroyo et al., 1997) and Mexican American adolescents (Arellano et al., 1997). Also, the rates reported in the current study were somewhat higher than estimated ranges in the general U.S. female population (i.e., 15% to 33%, (Briere & Runtz, 1988; Finkelhor et al., 1990; Saunders et al., 1992; Wyatt, 1985). However, this is likely due to this study’s use of a more robust multi-level measure of severity of sexual coercive and abusive experiences, the cutoff age of 18 year old, and the inclusion of similar aged peers as perpetrators in the definition of CSA. Nonetheless, 19% of women who experienced some form of sexual abuse fell into the most severe category of rape, and the average age at which the women said the first experienced their most severe level of abuse was 13 years old. This age is close to the results in the Romero et al. study in which the mean age which the women reported to first experience sexual abuse was 11 years old. In addition, a higher rate of women in this study reported more than one incident of the most severe type of abuse than in the Romero et al. study (47% versus 35%, respectively). These differences however, may be a result of this study’s different measure of sexually abusive experiences.

Similar to other studies with Latina samples, the results indicated that perpetrators of abuse were predominantly family members. However, this study also had a very high rate of peer sexual abuse (i.e., boyfriends and friends as perpetrators). When combined, these two categories accounted for 52% of the perpetrators. This finding highlights the importance of assessing for peer and dating partner perpetrated sexual abuse when assessing women’s overall histories of abuse. Moreover, the majority of perpetrators of rape experienced by the women before age 13 were friends and boyfriends. This finding was different from previous studies, but the authors are unaware of any other studies that have examined differences in perpetrators across severity of abuse events. Age and severity of abuse may be important factors when examining differences in perpetrators of abuse. Not including peer and dating-partner perpetrated abusive incidents in the assessment of childhood and adolescent sexual abuse may underestimate the prevalence of sexual abuse experiences.

As in the Romero et al. (1999) study, there was a high rate of non-disclosure of abuse events in this study (44% of those who reported an abusive event). In addition, the more severe the level of abuse, the less likely the women were to disclose the abuse event. In general, research has shown that Latina women are less likely to initiate formal mental health, help-seeking behavior (Padgett, Harman, Burns, & Schlesinger, 1998). The results from this study highlight the importance of conducting additional outreach to both English and Spanish speaking Latina women who may have been victims of abuse.

Mental health providers and abuse prevention research should take special precautions to address the cultural beliefs and issues specific to Latina women such as immigration status and fear of reporting family perpetrators of abuse. Although disclosure rates did not vary by generational status or language preference in this study, immigration issues may still be an important factor in whether or not Latina women disclose their abuse experiences. They may have fears of deportation and mistrust of law enforcement. Assessing immigration status must be done in an ethical manner with careful consideration for respondents’ confidentiality. The current study did not specifically ask about U.S. citizenship.

In addition, this study only measured generational status and language preference, and it did not include a specific measure of acculturation. While these items are components of most uni-dimensional measures of acculturation (Abraido-Lanza, Armbrister, Florez, & Aguirre, 2006), the exclusion of a specific measure of acculturation represents a limitation of the current study. For future research, the inclusion of a multidimensional acculturation measure would provide a more comprehensive understanding of the association between acculturation and experiences of CSA.

Despite the potential barriers, when the women did disclose their abuse experiences, three quarters reported feeling supported. People to whom the women said they most frequently disclosed their abuse were family members. This finding suggests that cultural values such as familismo may play an important role in facilitating social support. Mental health service providers who treat Latina survivors of abuse may want to highlight the importance of family and the potential benefits of gaining social support from trusted family members. However, mental health providers and researchers should be mindful of the complexity of issues and circumstances surrounding abuse. For example, family members and peers were also the most common perpetrators of violence. Unfortunately, this study did not ask the gender of the family members who perpetrated the abuse. Therefore it is possible that the abusers were male family members and the persons from whom the women sought support were their mothers or other female family members. Nonetheless, cultural issues such as respeto and familismo should be addressed when working with Latino/a populations.

This study is one of a few that has examined childhood and adolescent sexually coercive and abusive events among a community sample of English and Spanish speaking Latina women. Prevalence rates and characteristics of childhood sexual abuse among predominantly White and multi-ethnic samples of women were thought to be generalizable to Mexican American women, but had not been specifically assessed through quantitative studies with community samples of Latina women. Another strength of this study is that it used a multi-level measure of sexual abuse that has been shown to provide more information to researchers about women’s sexual abuse experiences (Roosa et al., 1998). Dichotomous measures that place CSA victims into either abuse or no-abuse categories underestimate the occurrence and uniqueness of various types of sexual abuse. By treating all abused women as a homogenous group, knowledge about differences in experiences and the circumstances surrounding CSA may be limited.

One of the limitations of this study as well as sexual abuse research in general, is the limited, fixed nature of sexual abuse measures. Most measures do not assess multiple experiences of abuse, chronic abuse, or multiple abusers and therefore may not reflect the true experience of sexual abuse for some women. For example, abuse by family members or relationship partners is not likely to be a single, discreet event. Victims of CSA have been shown to be at risk for re-victimization (Wyatt, Guthrie, & Notgrass, 1992). Furthermore, previous research has indicated that frequency and chronicity of the abuse can be important moderating factors in the relationship of sexual abuse to maladjustment (Browne & Finkelhor, 1986; Kendall-Tackett, Williams, & Finkelhor, 1993). The present study utilized a measure of the first experience of several types of sexually coercive and abusive events. Future research on sexual abuse would benefit from the development of a measure that can also assess the sum of abuse experiences.

The present study is also retrospective in nature, which can be confounded by potential memory bias. It has been shown that memory is subject to influence depending on many factors such as current emotional states and current life events (Ellis & Ashbrook, 1989). In addition, the present study only focused on the abuse experiences of women. Therefore, the results may not generalize to the experiences of male sexual abuse victims. Other research has focused specifically on men to determine how they are affected and if they are affected in similar ways as women (Dhaliwal, Gauzas, Antonowicz, & Ross, 1996). It is possible that CSA experiences are similar between young boys and girls because of the age and the family dynamics surrounding CSA. However, the experiences of CSA and the associated events such as physical violence and coercion may be very specific to women and may not occur in the same way for men.

Likewise, the results of this study may not be generalizable to other ethnic groups of women or even other Latina ethnic groups. This was a non-random, community sample of Latina women, many of whom were married, and some of whom were pregnant. However, it was the intention of this study to examine Latina women, specifically those of Mexican descent living in Southern California. The possibility of systematic sampling bias was reduced by recruiting participants at multiple sites that varied in size and location. Nevertheless, some caution is warranted in generalizing results from this sample.

Conclusions

Despite the limitations, this study provides important information on a relatively under-studied group of women. It replicates many of the previous findings from one other study of CSA in a community sample of Latina women (Romero et al., 1999), and it provides additional details on the severity of sexually coercive and abusive experiences. Peer and dating-partner violence emerged as an important factor to consider when assessing history of childhood and adolescent sexual abuse. In addition, results showed high rates of non-disclosure of abuse events. Disclosing abuse to a trusted family member may be beneficial since the majority of women who did disclose their abuse told family members and reported feeling supported.

Given the complexity and specificity of issues for Latinas, future research endeavors are needed that can address issues such as acculturation, citizenship, and cultural values in more depth with larger, cross-sectional as well as longitudinal study designs. This information is needed for the scientific study and treatment of the long-term psychological consequences of CSA in this growing population.