Introduction

Sexual Abuse, Methodological Difficulties, and Measurements

The World Health Organization (World Health Organization, 2010) defines sexual abuse (SA) as “Any sexual act, attempt to obtain a sexual act, unwanted sexual comments or advances, or acts to traffic or otherwise, directed against a person’s sexuality using coercion, by any person regardless of their relationship to the victim, in any setting, including but not limited to home and work” (p. 2). The need to define the concept has been widely recognized by leading researchers since the 1970s (Mathews & Collin-Vézina, 2019). World organizations, such as the World Health Organization and International Society for Prevention of Child Abuse and Neglect (2006), stated that the measurement of child abuse had until then been unsatisfactory, due to the many forms, consequences, and circumstances of violence, and encouraged those involved in examining child maltreatment to develop a common conceptual definition. Researchers argued that the concept of sexual abuse is extremely broad. Even the term “sexual abuse” is imprecise and can range from inappropriate suggestive speech to rape (Knudsen, 1988), which makes it rather complicated to operationalize the construct (Haugaard, 2000). There are also differences in the literature on how researchers define the upper age limit of childhood, which can vary from as young as 13 to 18 years old (Senn et al., 2008). The type of sexual act considered to be abusive can also differ across studies (DiLillo, 2001; Senn et al., 2008), with some being more restrictive (e.g., only penetrative contact) and some being less restrictive (e.g., non-contact acts, sexual requests, exhibitionism) (Swahnberg, 2003). Due to such differences in definition, significant variance can be recognized in findings regarding sexual abuse (Mathews & Collin-Vézina, 2019).

Despite these difficulties in assessing the phenomena, there are several scales currently in use in research and clinical settings (Bernstein et al., 1994, 2003; Hudson & McIntosh, 1981; Kooiman & Ouwehand, 2002; Marshall, 1992; McFarlane, et al., 1992; Sanders & Becker-Lausen, 1995; Straus, 1979; Swahnberg, 2003; Tolman, 1999). These scales measure different aspects of sexual abuse, varying on measured severity or target groups, with some focusing on intimate partner violence (Hudson & McIntosh, 1981), violence from men against women (Marshall, 1992), and physical and sexual violence against women (McFarlane, et al., 1992). Some focus on maltreatment in childhood and adolescence (Bernstein et al., 1994, 2003; Kooiman & Ouwehand, 2002; Sanders & Becker-Lausen, 1995); intrafamilial violence (Straus, 1979); psychological maltreatment of women (Tolman, 1999); emotional, physical, and psychological abuse among women (Swahnberg, 2003); and CSA history in adults (Lock et al., 2005). A majority of these measurement tools, including the Sexual Abuse History Questionnaire (SAHQ) (Leserman et al., 1995), were developed and validated on samples composed exclusively of women, with a small number of exceptions, where target groups were exclusively men (e.g., Swahnberg, 2011). Despite being a commonly used measure in clinical and research settings, the SAHQ has not yet been validated nor had its psychometric properties exhaustively investigated, with the exception of a limited number of studies, which included samples of women, men, and specific clinical populations (Swahnberg, 2003, 2011). A great advantage of the SAHQ (Leserman et al., 1995) is its exclusive focus on SA and its ability to measure revictimization, meaning experiencing sexual abuse more than once during one’s lifetime (Das & Otis, 2016). This is especially valuable since there is evidence that people who experienced abuse during childhood are more likely to become victims again than people who did not (Black et al., 2011; Das & Otis, 2016). However, the scale is only suitable for detecting more serious forms of sexual abuse (i.e., revealing genitals, threatening with sexual abuse, touching genitals, being forced to touch genitals, and being forced to have sex) (Leserman et al., 1995). This does not fully meet the current, internationally accepted definitions of sexual abuse (e.g., American Psychological Association (APA; Kazdin, 2000), Centers for Disease Control and Prevention (CDC, Leeb, 2008), or the World Health Organization (2010)) that include less severe or non-contact forms of sexual abuse as well (i.e., blackmail, psychological or physical intimidation, sexual humiliation, not being able to give informed consent—for instance, while being intoxicated, sleeping, or being mentally incapacitated or refusing to use safe sex practices).

To summarize, current research lacks a comprehensive measure that can handle all the constructs mentioned above, is fully aligned with the World Health Organization’s definition of forms of SA, and includes multiple aspects of the abusive event. Accordingly, we aimed to address these shortcomings by using a pre-established, data-driven practice (Bőthe, Koós et al., 2021) based on thematic analysis practices (Braun & Clarke, 2006) and examined the responses given to an open-ended question in the SAHQ (“Have you had any other unwanted sexual experiences not mentioned above? If yes, please specify”).

Sexual Abuse and Its Relationship with Other Sexuality-related Constructs

Decades of research suggest that sexual abuse history is associated with a number of short- and long-term psychological difficulties and mental health disorders, such as depression, anxiety, suicidality, posttraumatic stress symptoms, substance abuse, and sexual risk behaviors (Briere, 1992; Hailes et al., 2019; Hillberg et al., 2011; Lalor & McElvaney, 2010; Luster & Small, 1997; Neumann et al., 1996; Paloucci et al., 2001;). These interactions are compounded by the fact that adversities in childhood often co-occur and are often paired with maladaptive family functioning which poses a high risk of mental disorders (Kessler et al., 2010). Sexual abuse history is a risk factor for sexual function problems and engaging in unrestricted sexual behaviors (Niehaus et al., 2010; Senn et al., 2007) and is associated with having more sexual partners (Merrill et al., 2003), a greater likelihood of having sex with someone they just met, earlier age at first intercourse, and higher frequency of sexually transmitted infection (STI) diagnoses (Walser & Kern, 1996).

Sexual abuse history has also been associated with difficulties in adult interpersonal functioning, including intimate relationships, which are often characterized by low satisfaction, more conflicts (Testa et al., 2005), problems with physical contact, poor communication, and lower levels of trust (Nielsen et al., 2018). The traumagenic model, developed by Finkelhor and Browne (1985), provides a satisfactory explanation for the association with interpersonal problems. The model lists four traumagenic dynamics: traumatic sexualization, betrayal, powerlessness, and stigmatization, which “alter children’s cognitive and emotional orientation to the world, and create trauma by distorting children’s self-concept, world view, and affective capacities” (p. 531). It is argued that out of the four traumagenic dynamics, betrayal seems to have the most influence on later interpersonal problems. The sense of loss and betrayal experienced in childhood due to being abused can carry on into later relationships, which results in suspicion, avoidance, or isolation from intimate relationships and difficulties in maintaining healthy intimate relationships (Davis & Petretic-Jackson, 2000).

The majority of clinical trials investigating the effects of intrafamilial or extrafamilial sexual abuse report sexual problems among survivors (Browne & Finkelhor, 1986; Randolph & Reddy, 2006). Women with a history of sexual abuse tend to experience dissociative symptoms and difficulties with sexual function more frequently than their non-abused counterparts (Bird et al., 2014). Among sexual dysfunctions, problems with sexual arousal appear to be the most common in women with a sexual abuse history (Lewis et al., 2010). Victimized individuals often experience aversion to sex, difficulties with orgasms, cases of vaginismus, or an overall negative attitude toward their sexuality (Finkelhor & Browne, 1985). Reported problems also include experiences of sexual guilt and dissatisfaction in sexual relationships and anxiety in relation to sexual activities (Browne & Finkelhor, 1986). Studies conducted predominantly among women showed positive associations with a lack of interest and avoidance regarding sexual activities and younger age at first voluntary intercourse (Noll et al., 2003; Randolph & Reddy, 2006).

Several theories have been proposed to explain the association between sexual abuse and compulsive sexual behavior (Alexander, 1992; Finkelhor & Browne, 1985; Gold & Heffner, 1998; Katehakis, 2009). According to the ICD-11 diagnostic criteria (World Health Organization, 2022), compulsive sexual behavior disorder (CSBD) is categorized as an impulse control disorder and characterized by a persistent pattern of failure in controlling intense sexual urges and behaviors to a point where these activities become the central focus of a person’s life, while other important personal needs, interests, activities, or responsibilities are neglected. The person continues to engage in these sexual behaviors, despite experiencing significant distress or impairment in important areas of personal and interpersonal functioning and little or no satisfaction (Bőthe et al., 2022; Gola et al., 2022; Sassover & Weinstein, 2020; World Health Organization, 2022). Survivors of sexual abuse often develop maladaptive schemas around sexuality that influence their sexual behaviors or beliefs, which can lead to repetitive or compulsive sexual behaviors (Finkelhor & Browne, 1985; Niehaus et al., 2010; Pachankis et al., 2014; Slavin, Blycker et al., 2020; Slavin, Scoglio et al., 2020). These traumatic experiences in childhood can affect the development of the prefrontal cortex and therefore decision-making, impulse control, and mood regulation among other functions (Katehakis, 2009).

Some have also argued that individuals who were overpowered sexually as children may use sexuality as a means of taking back control (Gold & Heffner, 1998), while some theorists emphasize the importance of attachment and the internal working models based on the relationship with caregivers, which influence self-concept and therefore sexual behaviors later on (Alexander, 1992). Therefore, survivors of sexual abuse may engage in compulsive sexual behaviors, including problematic pornography use, more frequently. Scientific evidence about the relationship between sexual abuse and problematic pornography use (PPU) is still limited. In some cases, more than 80% of people meeting the criteria for CSBD have also reported excessive pornography use (Kafka, 2010; Reid et al., 2012). This pattern was also observed in clinical samples, where reported PPU was around 50% (Briken et al., 2007; Reid et al., 2009). These findings indicate that PPU is a frequent manifestation of CSBD (Bőthe et al., 2019), which combined with the similarities between the two phenomena (such as impulsivity, compulsivity) (Bőthe et al., 2019) may suggest the existence of similar underlying neurological mechanisms (Katehakis, 2009). Therefore, similar relationship patterns might be expected between SA and PPU as between SA and CSBD.

The Aim of the Current Study

The aim of the present study was to investigate the psychometric properties of the SAHQ in a large non-clinical sample and explore sexual abuse as a complex phenomenon through clustering the answers following pre-established practice, based on thematic analysis practices (Braun & Clarke, 2006). Furthermore, we aimed to explore the relationships between sexual abuse and compulsive sexual behaviors, relationship satisfaction, sexual function, problematic pornography use, and other sexuality-related characteristics (i.e., number of sexual partners, number of casual sexual partners, frequency of having sex with a partner, frequency of having sex with a casual partner).

  • Hypothesis 1. We hypothesized that CSA and AASA would be associated positively.

  • Hypothesis 2. We hypothesized that both AASA and CSA would positively relate to compulsive sexual behaviors (CSBs).

  • Hypothesis 3. We hypothesized that both AASA and CSA would positively relate to problematic pornography consumption.

  • Hypothesis 4. We hypothesized that both AASA and CSA would negatively relate to sexual function.

  • Hypothesis 5. We hypothesized that both AASA and CSA would negatively relate to relationship satisfaction.

  • Hypothesis 6. We hypothesized that both AASA and CSA would positively relate to the number of sexual partners (either long-term or casual).

  • Hypothesis 7. We hypothesized that both AASA and CSA would positively relate to the frequency of having sex (either with a long-term partner or with a casual partner).

Method

Procedure and Participants

The present study was part of a larger project. Subsamples from this dataset were used in previously published studies (all previously published studies and included variables can be found at OSF; masked for review). Data were collected via an online questionnaire that was promoted on one of the largest news portals in Hungary. We used pre-established rules of thumb to estimate the minimum required sample size (VanVoorhis et al., 2007) and had no pre-established rule for stopping the data collection. Data collection was conducted in January 2017. Study participants must be aged 18 years old or older. Participants received detailed information about the aims of the study (i.e., investigation of sexual habits and behaviors), and they were assured of anonymity and confidentiality. Informed consent was then obtained from all participants included in the study. Participants were compensated (they had a chance to win one of three tablets). The survey took approximately 30 min to complete. For further information on the original sample and exclusion criteria, see Bőthe et al. (2019).

Our total sample thus consisted of 24,715 individuals. The sample was separated into two non-overlapping subsamples randomly while preserving similarity across various demographic domains (e.g., gender, sexual orientation, socio-economic status, education, marital status). Sample 1 included 12,251 individuals (men = 67.9%, women = 31.2%, gender-diverse individuals = 0.8%), whereas sample 2 included 12,464 individuals (men = 68.8%, women = 30.2%, gender-diverse individuals = 0.9%). The detailed sociodemographic characteristics of the sample are shown in Appendix A. 

The study was approved by the Institutional Review Board of the research team’s university and conducted following the Declaration of Helsinki.

Measures

Sexual Abuse History Questionnaire (Leserman et al., 1995)

The SAHQ is a 12-item scale that assesses sexual abuse in childhood (13 or younger) and adulthood (14 or older). The scale includes items regarding abuse with physical contact (e.g., “Has anyone ever touched the sex organs of your body when you did not want this?”), without physical contact (e.g., “Has anyone ever exposed the sex organs of their body to you when you did not want it?”), and forced intercourse (“Has anyone ever forced you to have sex when you did not want this?”) both in childhood and adulthood. Participants answered these questions with a simple “yes” or “no.” Higher scores on the CSA or AASA scale indicated the accumulation of multiple forms of sexual abuse. The sample distribution regarding the type of abuse indicated can be seen in Appendix B. The scale includes an additional open-ended question assessing potential abusive or unwanted sexual experiences (“Have you had any other unwanted sexual experiences not mentioned above? If yes, please specify”). The scale was originally developed with individuals seeking treatment in a gastroenterology clinic. The measure has demonstrated adequate reliability and validity in previous studies (Leserman et al., 1995). The SAHQ was translated as part of a larger project mentioned above following a pre-established translation and back-translation protocol (Beaton et al., 2000).

Problematic Pornography Consumption Scale (Bőthe, Tóth-Király et al., 2018)

The PPCS consists of six factors based on the six-component addiction model (Griffiths, 2005) and includes three items per factor: salience (e.g., “I continually planned when to watch porn”), tolerance (e.g., “I felt that I needed more and more porn in order to satisfy my needs”), mood modification (e.g., “Watching porn got rid of my negative feelings”), relapse (e.g., “I resisted watching porn for only a little while before I relapsed”), withdrawal (e.g., “I became stressed when something prevented me from watching porn”), and conflict (e.g., “I neglected other leisure activities as a result of watching porn”). The reliability of the scale was excellent in the original validation study (α = 0.93) (Bőthe, Tóth-Király et al., 2018). Participants indicated their answers regarding the past 6 months on a 7-point Likert scale (1 = never; 7 = very often).

Sexual Functioning Scale (Burwell et al., 2006; Sherbourne, 1992)

Sexual functioning was assessed with four questions related to different aspects of sexual functioning: lack of interest in sexual activities, difficulty in becoming sexually aroused, difficulty in achieving orgasm, and difficulty in enjoying sex. Respondents indicated their level of problems on each dimension on a 4-point scale (1 = “not a problem”; 4 = “much of a problem”). A higher score indicated greater difficulties. The scale demonstrated adequate reliability in previous studies (e.g., Broeckel et al., 2002; Zebrack et al., 2010).

Hypersexual Behavior Inventory (Reid et al., 2011)

The Hypersexual Behavior Inventory (HBI) is a 19-item scale which assesses hypersexual behavior based on Kafka’s (2010) proposal for the DSM-V (American Psychological Association, 2013). It measures hypersexuality (or compulsive sexual behavior, World Health Organization, 2022) via three dimensions: coping, control, and consequences. The coping factor consists of seven items and refers to sex and sexual behaviors as responses to negative emotions or distress (e.g., “I use sex as a way to try to help myself deal with my problems.”). The control factor consists of eight items and refers to difficulties in managing sexual urges, behaviors, and fantasies (e.g., “Even though my sexual behavior is irresponsible or reckless, I find it difficult to stop.”). Finally, the consequence factor includes four items and refers to the interference of sexual thoughts and activities with education, occupational duties, or interpersonal relationships (e.g., “I sacrifice things I really want in life in order to be sexual”). Participants indicated their answers on a 5-point Likert scale (1 = never; 5 = very often). The overall reliability for the scale was α = 0.95 in the original study (Reid et al., 2011) and in subsequent studies as well (e.g., Bőthe, Kovács et al., 2018).

Relationship Satisfaction

Relationship satisfaction was assessed using a single-item measure (Fülöp et al., 2020) based on the original Relationship Assessment Scale (RAS, Hendrick et al., 1998). The RAS-1 showed similar psychometric qualities as the original RAS. Therefore, the RAS-1 was used in the present study. Respondents indicated answers on a 5-point scale (1 = not satisfied; 5 = very satisfied). Only participants who were in a relationship at the time of the data collection completed this measurement.

Sexuality-related Questions

Following the standard assessment of demographic characteristics (gender, age, sexual orientation, relationship status), additional sexuality-related information (Bőthe, Kovács et al., 2018) was asked from the participants, such as the number of sexual partners in one’s lifetime (16-point scale, 1 = no partner, 16 = more than 50 partners), the number of casual partners in one’s lifetime (16-point scale, 1 = 0 partner, 16 = more than 50 partners), frequency of sex with a partner in the last year (10-point scale, 1 = never, 10 = 6 or 7 times a week), and frequency of sex with casual partner in the last year (10-point scale, 1 = never, 10 = 6 or 7 times a week).

The descriptive statistics of the aforementioned measurements are included in Table 1 under the results.

Table 1 Descriptive statistics, normality indices, and psychometric properties of the scales
Full size table

Statistical Analysis

For cleaning and organizing data, conducting descriptive analyses, and performing correlations, the IBM SPSS 25 (IBM Corp, 2017) software was used, while all the other statistical analyses were conducted using Mplus 8 (Muthén & Muthén, 1998–2018).

All analyses were prespecified before examining the data, following the data collection. Two subsamples were created to perform both EFA and CFA, since there was no pre-established factor structure regarding the scale. Exploratory factor analysis (EFA) was conducted to examine dimensions of the SAHQ in sample 1 (n = 12,251) from one- to three-factor solutions regarding AASA and CSA separately. The rotated solutions (oblique rotation of Geomin) with standard errors were obtained for the factors. Weighted least square mean and variance (WLSMV) estimators were used due to having categorical variables. The goodness of fit was assessed (Schermelleh-Engel et al., 2003) by commonly used goodness-of-fit indices (Brown, 2015): the root-mean-square error of approximation (RMSEA; ≤ 0.06 for good, ≤ 0.08 for acceptable), the Tucker-Lewis index (TLI; ≥ 0.95 for good, ≥ 0.90 for acceptable), and the comparative fit index (CFI; ≥ 0.95 for good, ≥ 0.90 for acceptable) with 90% confidence intervals. Two reliability indices, Cronbach’s alpha (Nunnally, 1978) and composite reliability (CR) index, were calculated to assess internal consistency, due to Cronbach’s alpha’s potentially decreased efficiency (e.g., McNeish, 2018; Schmitt, 1996; Yang & Green, 2011). The CR index was calculated by the formula of Raykov (1997). Confirmatory factor analysis (CFA) was then conducted in sample 2 (n = 12,464) to test the previously identified model using the mean- and variance-adjusted weighted least square (WLSMV) estimators. The observed fit indices and reliability indices were the same as in the case of the EFA.

The associations between the SAHQ scores and the sexuality-related questions (i.e., number of sexual partners in one’s lifetime, number of casual partners in one’s lifetime, frequency of sex with a partner in the last year, frequency of sex with casual partner in the last year), problematic pornography consumption (PPCS), sexual dysfunction (SFS), compulsive sexual behavior (CSB), and relationship satisfaction (RAS-1) were examined using Spearman correlations.

Additional forms of abuse were assessed via an open-ended question (“Have you had any other unwanted sexual experiences not mentioned above? If yes, please specify”). As Appendix B shows, a total of 2309 individuals indicated the occurrence of any other unwanted sexual event in their lifetime. Of these, 1595 also provided a text response detailing the event, which was used in the analysis. After exporting the data (n = 1595), responses that were not relevant or had already been assessed by an item of the SAHQ were excluded in the first stage (n = 772). In the next stage of data cleaning, responses, whose context did not clearly indicate the non-consensual traumatic nature of the event (e.g., touching of thighs), were excluded from the analysis (n = 248). The authors would like to acknowledge the potentially traumatic nature of these unpleasant experiences; however, interpreting them as consent-violating would have been arbitrary. During the coding process, six more responses were excluded that fit one of the two exclusion criteria described above, as a result of consensus. In the final analysis, a total of 569 responses were analyzed (see Fig. 1). Following the first review of the answers, eight categories were identified following a previously established procedure (Bőthe, Koós et al., 2021) based on thematic analysis practices (Braun & Clarke, 2006). The answers were classified by two independent coders into these previously established categories, which corresponded to the most prevalent forms of sexual abuse that had been identified. The two coders were psychology students in their bachelor’s and master’s degree. Any conflicts that arose during the coding process were resolved on two occasions with the involvement of the supervisor of the students and a certified sexual psychologist.

Fig. 1
figure 1

Flow chart of sampling regarding additional SA experiences

Full size image

Results

Results of the Exploratory Factor Analysis in Sample 1

CSA and AASA were examined separately, not considering them as two factors in one scale. To identify the best factor solutions for both CSA and AASA, three models were tested for each, out of which only the one-factor models showed acceptable fit to the data both in childhood (CFI = 0.996; TLI = 0.991; RMSEA = 0.027 [90% CI = 0.020–0.035]) and adolescence/adulthood (CFI = 0.995; TLI = 0.990; RMSEA = 0.034 [90% CI = 0.026–0.041]). All items had strong factor loadings in the one-factor structure model (see Table 2). Two reliability indices were calculated to examine the internal consistency of the identified components (CSA, AASA, AASA without item 1). The Cronbach’s alphas were somewhat lower than expected (0.60–0.68), but the composite reliability index (0.90–0.90) showed excellent values (see Table 1).

Table 2 Factor loadings in sample 1 (EFA) and sample 2 (CFA)
Full size table

Results of the Confirmatory Factor Analysis in Sample 2

As the next step in the analysis, we conducted CFAs to further test the construct validity of the three components (i.e., CSA, AASA). The models showed acceptable fit to the data both in childhood (CFI = 0.995; TLI = 0.991; RMSEA = 0.031 [90% CI = 0.024–0.039]) and adolescence/adulthood (CFI = 0.995; TLI = 0.990; RMSEA = 0.037 [90% CI = 0.030–0.044]). Reliability indices were calculated (see Table 1). Similarly to sample 1, the Cronbach’s alpha indicators were lower (0.66–0.70). However, the composite reliability indices were excellent (0.90–0.95). Results of the CFAs can be seen in Table 2.

Sexual Abuse and Its Correlates

Associations between AASA, CSA, and sexuality-related measures were examined in both subsamples (see Table 3). AASA showed a weak positive association with compulsive sexual behavior, difficulties with sexual function, number of sexual partners, and casual sexual partners, and frequency of having sex with a partner in both subsamples. It correlated weakly and negatively with PPU.

Table 3 Bivariate correlations between sexual abuse and other sexuality-related constructs
Full size table

CSA was weakly and positively associated with compulsive sexual behavior, difficulties with sexual functioning, number of sexual partners, and casual sexual partners in both subsamples. Frequency of having sex with a partner showed a weak positive correlation to CSA in sample 1; however, it had a negative one in sample 2. CSA was also negatively associated with PPU in both subsamples and in sample 2 with relationship satisfaction. CSA and AASA correlated positively in both subsamples, although these were weak to moderate associations.

Additional Forms of Sexual Abuse Identified via Open-ended Questions

After data cleaning, all valid text responses were grouped into the following categories: groping (e.g., “as a child I was lured away on the beach and groped by an old man”), non-physical coercion (e.g., “My partner made me have sex with him even when I didn't want to. With emotional pressure, persuasion”), lack of consent due to altered consciousness (e.g., “having sex in an almost unconscious state under the influence of alcohol”), verbal abuse (e.g., “My father's friend made innuendos about my sexual maturation, sexual propositions, sexual fantasies”), physical harm in the context of consensual sexual activity (e.g., “I slept with my best friend, but the next day I was hospitalized for internal injuries from bruising inside my genitals”), consent regarding sexual health and the reproductive system (e.g., “The partner refused to use a condom, forcing me to accept his decision”), breach of ongoing consent (e.g., “unexpected, unwanted events during an otherwise consensual encounter”), and other (e.g., “My father regularly watched me bathe and masturbated”). Depending on their content, responses could fall into several categories. Altogether 60% (n = 341) of the participants reported groping, 17% (n = 97) have ever experienced non-physical coercion, 14% (n = 77) had experience with not being able to provide consent due to altered consciousness, and 11% (n = 63) were ever affected by sexually loaded verbal abuse. About 5% (n = 29) of participants experienced the breach of ongoing consent regarding any sexual activity, 3% (n = 16) of participants indicated that they were physically harmed in the context of consensual sexual activity, and 2% (n = 9) have experienced abuse of consent regarding their sexual health and reproductive system. Finally, 10% (n = 55) of participants were grouped into the category “other.” Participants who were grouped into this category have additionally indicated some sort of physical abuse, physical restraint (e.g., being locked into a car or a room during the event of the abuse), or indicated forms of abuse which the other groups did not cover.

Discussion

The importance of the present study lies in thoroughly investigating the psychometric properties of a scale commonly used in clinical and research settings and highlighting the shortcomings of the scale, thus serving as a starting point for the development of a measurement tool covering all aspects of SA. Specifically, the aim of the present study was to investigate the psychometric properties of the SAHQ (Leserman et al., 1995), examine the relationship between sexual abuse and other sexuality-related constructs, and identify additional forms of abuse that the scale does not measure. The results of the present study showed that the SAHQ has adequate psychometric properties in terms of internal consistency, composite reliability, and structural validity. However, the criterion validity of the scale warrants further investigation, as only weak correlations were observed, which could be due to the large sample size with other sexuality-related constructs. Therefore, further research is needed regarding the psychometric qualities of the scale. Additionally, there are certain forms of sexual abuse that cannot be properly addressed through this scale, suggesting that the SAHQ might not cover sexual abuse in its full complexity.

Both the exploratory and confirmatory analyses suggested a one-factor structure both in childhood and adulthood on two independent samples. The proposed models showed an adequate fit to the data in both subsamples. However, the Cronbach’s alphas were lower than expected in all cases (i.e., CSA, AASA). These lower values are presumably due to the fact that the broad concept of SA was measured with relatively few items (Cortina, 1993). Therefore, the composite reliability index (Raykov, 1997) was calculated to further investigate these lower values and suggested acceptable reliability in all samples.

Sexual Abuse and Its Associations with Sexuality-related Constructs

CSA and AASA were associated moderately and positively in both subsamples. These findings support the previously established concept that sexual abuse in childhood increases the likelihood of later revictimization (Black et al., 2011; Das & Otis, 2016). The majority of supporting results were obtained predominantly from samples including only women; however, some studies have observed that men and adolescents may be more likely to be revictimized (Classen et al., 2005; Desai et al., 2002). Some have suggested that the impairment of risk perception in victimized individuals may also explain subsequent revictimization, which may be the result of psychological disturbances (e.g., PTSD, dissociative symptoms) as a consequence of CSA (Arata, 2002). It is important to note, however, that the relationship between CSA and AASA can also be conceptualized differently, as childhood abuse can also lead to hypo-sexuality or sexual aversion (Wohl & Kirschen, 2018). In addition to physical symptoms of significant pain and distress, lower sexual desire or sexual aversion can be explained by psychological consequences such as intense fear or panic, loss of control, disgust, or shame as a result of the abuse and sometimes also by severe symptoms such as PTSD-associated flashbacks (Bohus et al, 2013). According to these findings, we conclude that our hypothesis regarding the relationship between CSA and AASA (H1) was fully supported.

CSB was positively and moderately associated with CSA and AASA in both subsamples. These findings are consistent with previous literature suggesting that survivors of SA often develop maladaptive schemas around sexuality, which may result in repetitive or compulsive sexual behavior (Finkelhor & Browne, 1985; Niehaus et al., 2010; Pachankis et al., 2014; Slavin, Blycker et al., 2020; Slavin, Scoglio et al., 2020; Vaillancourt-Morel et al., 2015). Therefore, we conclude that our hypothesis regarding the relationship between SA and CSB (H2) was fully supported.

Problematic pornography use was negatively associated with SA in both subsamples. This result is surprising, as several studies have previously found that PPU is a frequent manifestation of CSB and therefore is associated with sexual abuse and maladaptive sexual functioning (Bőthe et al., 2019; Briken et al., 2007; Reid et al., 2009). However, this negative relationship can be conceptualized along the lines of sexual aversion. As mentioned above, besides hypersexuality, hypo-sexuality and aversion to sexual material can also appear as a consequence of sexual abuse (Vaillancourt-Morel et al., 2015; Wohl & Kirschen, 2018). Accordingly, individuals who have experienced sexual abuse in their lifetime may be more inclined to avoid potentially triggering impulses, such as pornographic material. Given that the relationship between the two constructs is under-researched and may be influenced by several latent factors, further research would be needed. In conclusion, our hypothesis regarding the relationship of SA and problematic pornography consumption (H3) was not supported.

Difficulties with sexual function were positively associated with CSA and AASA in both subsamples. The moderate but significant positive association between sexual dysfunction and sexual abuse is also consistent with prior results showing sexual dysfunction among women with SA history (Pulverman et al., 2018) and thus is often discussed as a risk factor (Pulverman & Meston, 2019). In conclusion, our hypothesis regarding the relationship of SA and difficulties with sexual function (H4) was supported.

Relationship satisfaction was weakly and negatively associated with CSA. However, this correlation was only present in sample 2. This result is consistent with previous literature, suggesting that sexual abuse may be associated with later relationship dissatisfaction (Friesen et al., 2009; Knapp et al., 2017; Lassri et al., 2018; Watson & Halford, 2010). The exclusive relationship between relationship satisfaction and CSA may be explained by attachment styles formed in childhood. Consistently, children and adolescents with a CSA history are more likely to be classified as having preoccupied and disorganized attachment styles (Ensink et al., 2020). Even in adulthood, a majority of CSA survivors experience considerable disruptions in attachment security (Labadie et al., 2018) and report higher levels of attachment anxiety and avoidance (Kwako et al., 2010). Consequently, it is possible that relationship satisfaction in adulthood has a deeper relation to CSA, as it is the relationship patterns developed in early ages that influence our later relationship patterns (Fonagy & Luyten, 2009). In conclusion, our hypothesis regarding the relationship of SA and relationship satisfaction (H5) was partially supported.

Both the number of sexual partners and the number of casual sexual partners showed a positive, moderate association with CSA and AASA in both subsamples. Frequency of having sex with a partner was associated positively but moderately with AASA in both subsamples, also with CSA in sample 1, and negatively with CSA in sample 2. Frequency of having sex with a casual partner showed no associations with either CSA or AASA. The higher number of both long-term and casual sexual partners may be explained by the theories presented in the “Introduction” section, which highlight, for example, maladaptive sexual schemas as a result of abuse (Finkelhor & Browne, 1985; Niehaus et al., 2010; Pachankis et al., 2014), the impact on attachment styles (Alexander, 1992), or the desire for taking back control (Gold & Heffner, 1998). These maladaptive sexual schemas may influence later sexual behaviors or beliefs and sexual behavior, leading to more sexual partners or more frequent sexual activity. Sexual behavior may also be influenced by attachment needs, such as reassurance or intimacy, overshadowing personal needs, including pleasure or sexual exploration, for example, frequently initiating sexual activities to avoid rejection in the case of an anxious attachment style (Labadie et al., 2018). In light of these findings, our hypothesis regarding the association between sexual abuse and the increased number of sexual partners (H6) was fully supported. However, our hypothesis regarding SA and increased frequency of sexual activities (H7) was only partially supported. The ambiguous results regarding the frequency of sexual activities indicate the necessity of further examination regarding this relationship.

Additional Types of Sexual Abuse

Additional types of sexual abuse were identified for the purpose of assessing sexual abuse experiences in a more complex way following a pre-established practice (Bőthe, Tóth-Király et al., 2021). As previously described, the SAHQ assesses more severe forms of sexual abuse, such as those involving physical touch of the genitals or specifically involving coercion to perform sexual acts. While these forms of sexual abuse fall within the broad definition of sexual abuse by the World Health Organization (2010) or the American Psychological Association (Kazdin, 2000), they do not cover all aspects of it. Consequently, the following additional categories have been identified:

  1. (i)

    Groping: Involuntary or non-consensual sexual advances involving touching, not necessarily only on the genitals (e.g., unsolicited kissing, touching, or rubbing against other parts of the body)

  2. (ii)

    Non-physical coercion: Any overt or covert emotional, verbal or non-verbal manipulation, blackmail, or abuse of power of any kind (e.g., intercourse with a partner for the sole purpose of avoiding a later argument or to keep calm)

  3. (iii)

    Lack of consent due to altered consciousness: Any sexual activity in a state that differs from the normal waking state, and it is assumed that the person concerned could not give full consensual consent (e.g., asleep, under the influence of alcohol or drugs)

  4. (iv)

    Verbal abuse: Any and all verbal and sexually charged advances that the person experiences as threatening, uncomfortable, or manipulative (e.g., verbal harassment at work, advances by a relative or older person) and which are not necessarily followed by any acts of sexual nature

  5. (v)

    Physical harm in the context of consensual sexual activity: Intercourse (that may be consensual) that results in unwanted pain or internal/external physical injury in the short or long term (e.g., complications) and causes distress to the respondent

  6. (vi)

    Consent regarding sexual health and the reproductive system: Any act where the perpetrator, despite a promise or a request from the partner, does not use a contraceptive technique that would avoid potential physical harm (unwanted pregnancy, transmission of STI) (e.g., non-consensual condom use, concealment of disease)

  7. (vii)

    Breach of ongoing consent: A person engaging in consensual sexual intercourse, during which they are progressively confronted with activities to which they have not consented (e.g., during sexual intercourse, partner asks them to do things that are outside their comfort zone and is experienced as traumatic)

An additional category was created in order to group those answers that did not fit any of the categories mentioned above.

There is little to no precedent in the literature for detailing forms of sexual abuse or what people tend to experience as traumatic or abusive in qualitative studies. However, there are some distinctions between contact and non-contact forms of abuse (Aebi et al., 2015; Collings, 1995; Jumper, 1995; Landolt et al., 2016; Witt et al., 2019). Similarly, the World Health Organization and International Society for Prevention of Child Abuse and Neglect (2010) or the Center for Disease Control and Prevention also differentiates between non-contact sexual abuse (unwanted exposure to sexual acts, such as pornography, voyeurism, exhibitionism), sexual harassment (e.g., comments or unwanted attention, prostitution or sexual trafficking), abusive sexual contact (e.g., intentional touching either directly or through clothing, not involving penetration, e.g., on the genitalia, anus, groin, breast, inner thigh, or buttocks), and sexual acts that include contact via penetration (Leeb, 2008). Our categories show a high degree of consistency with the World Health Organization and International Society for Prevention of Child Abuse and Neglect (2010) and CDC (2008) definitions and categories. Some measurement tools have already adopted different categories of sexual abuse (e.g., Swahnberg et al., 2003; Swahnberg et al., 2011). However, in the future, it may be worth developing a measurement tool along these identified categories that includes all potential forms of abuse.

Limitations and Future Studies

The present study was cross-sectional, limiting causal inferences. The self-report methods used may have biases (e.g., underreporting and overreporting). Furthermore, the SA total score does not fully capture the severity of the abuse, only the accumulation of types of SA, not culpability or proximities (e.g., how close the perpetrator was). Even though the data were not representative of the population (e.g., it excluded people without internet access, no interest in reading news websites, or this particular website), our sample was relatively diverse in terms of gender distribution, age, sexual orientation, and other sociodemographic characteristics. Future studies are recommended to further validate these measurements in clinical and representative samples and in culturally diverse (e.g., international) settings. Cross-cultural data could assist us in eliminating measurement issues regarding sexual health, by providing well-validated instruments and insights to improve theoretical understanding of various sexuality-related contracts (Bőthe, Tóth-Király et al., 2021). These results could also assist us in developing a measurement tool that can accurately and comprehensively measure sexual abuse.

Conclusions and Implications

The current study suggests that the SAHQ has adequate psychometric properties in terms of internal consistency, composite reliability, and structural validity, but the criterion validity of the scale requires further studies. Our findings also suggested that the SAHQ might not measure sexual abuse in its full complexity. Our results can help develop more comprehensive measurement tools and clinical practice by providing additional aspects of sexual abuse experiences that might not have been considered previously.

Social policy implications may include that measuring both CSA and AASA poses significant methodological difficulties, but identifying these experiences at large is inevitable and necessary. Measures that were developed a few decades ago may not keep pace with the ever-expanding, evolving definitions and experiences of sexual abuse that affect a large part of society. The present research may point to the need to apply these broader definitions of sexual abuse to acquire more diverse and realistic results. These implications are also of great importance when it comes to designing interventions or treatment for sexual abuse survivors, as they can result in more personalized, inclusive, and effective intervention approaches.