Alcohol, drug use and experiences of sexual violence victimisation among first-year college students in Ireland

ABSTRACT

Sexual violence victimisation is common in college and first year presents a significant period of risk due to the navigation of alcohol, drugs and engaging in sexual intimacy. Cross-sectional data were analysed from 1,778 first-year college students, aged between 18 and 25 years. Experience of different forms of sexual violence victimisation and perpetrator tactics used were explored, along with the reported alcohol and drug use habits of victims. Results found that sexual violence through incapacitation was the most frequently reported tactic among all students. Hazardous alcohol consumption was significantly associated with higher rates of experience of all forms of sexual violence since beginning college among female students, and significantly associated with an increased likelihood of experiencing unwanted sexual touching for male students. There were gender differences in the associations between drug types and forms of sexual violence experienced. All drugs assessed predicted an increased risk of experiencing rape among female students, while cocaine, ecstasy and ketamine were associated with an increased risk of rape among male students. Further investigation is required into the context and environments in which different drug types are taken. The findings provide a strong rationale for including alcohol and drug use education as part of future sexual health and consent education programmes for college students.

PRACTICE IMPACT STATEMENT

This study has implications for the promotion of positive sexual health and the reduction of sexual violence victimisation among college students. In order to understand the forces at play regarding sexual consent practices among students it is necessary to recognise the contexts in which these incidents occur. This research informs third-level institutions (i.e. universities, third-level colleges) on the importance of including alcohol and drug use as part of effective sexual consent education programming due to the frequent overlap of substance use and sexual experiences among this particular cohort.

KEYWORDS:

• Sexual violence
• victimisation
• alcohol
• drugs
• college students

Introduction

Entering third-level education represents a significant time of transition in a young person’s life, both academically and personally as they navigate a new environment. The independence associated with starting college is combined with a notable shift in personal responsibilities, a sudden change in social settings, and a drop in institutional supports (Evans et al., 2009; Morosanu et al., 2010; Schulenberg et al., 2004). For the majority, this transition occurs during what can be referred to as emerging adulthood between the ages of 18 and 25 years, a period distinct from both adolescence and adulthood, characterised by exploration and change for many young people (Arnett & Tanner, 2006). This period of exploration often represents a peak in levels of a number of risk behaviours (Arnett & Tanner, 2006), and most young adults are in an early phase of managing the use of alcohol, drugs (Hingson et al., 2017; Li et al., 2016), and how they engage in sexual intimacy (Claxton & van Dulmen, 2013; Downing-Matibag & Geisinger, 2009). Many first-year students have previously received inadequate sexual health education (Dawson et al., 2019; Shapiro & Brown, 2018; Willis et al., 2019), and in combination with a likely increase in exposure to alcohol and drugs this can contribute to a heightened risk of experiencing sexual violence among first-year students in the first few weeks of their first semester in particular, sometimes referred to as the “red zone”(Cranney, 2015; Flack et al., 2008). Research carried out in third-level institutions has shown that a significant number of students experience sexual violence of some kind during their time in college (Cantor et al., 2020; Muehlenhard et al., 2016), with as many as 1 in 5 female students reporting sexual assault during that time (Muehlenhard et al., 2017). In the UK a study of college students reported that one in three respondents had experienced unwanted sexual contact, while 12% had experienced unwanted sexual intercourse (NUS, 2019). In 2013, a study of third-level students in Ireland found that 16% of students reported having experienced some form of unwanted sexual experience while in their current educational institution (USI, 2013).

Extent of sexual violence in colleges

Unwanted sexual contact and sexual coercion are the most common forms of sexual violence reported on college campuses, followed by incapacitated rape, and completed or attempted forcible rape (Fedina et al., 2018). The campus climate survey methodology developed in the U.S. has provided a major resource for measuring experiences of sexual violence and harassment among college students (Wood et al., 2017). The AAU survey showed that 26% of undergraduate female students described at least one occasion of non-consensual sexual contact by force or while incapacitated and unable to give consent (Cantor et al., 2020; Howard et al., 2019), with 7% of male, and 23% of gender non-conforming students reporting similar experiences (Cantor et al., 2020). Instances of sexual violence are particularly common in first year, with 16.1% of young women reporting non-consensual sexual contact by physical force or inability to consent during the first year of college (Carey et al., 2015). One comparative study in the US found an increase in completed sexual violence victimisation since age 14 reported in 2015 (23.9%) in comparison to 1985 (15.8%) (Koss et al., 2022).

The role of alcohol

Most sexual assaults on college campuses involve alcohol consumption by the victim and/or perpetrator (Abbey, 2011; de Bruijn & de Graaf, 2016; Messman-Moore et al., 2008; Shorey et al., 2011; Smith & Wessel, 2011). Although high levels of alcohol consumption are not a prerequisite to sexual violence perpetration or victimisation, the common co-occurrence implies that alcohol may play a role in some incidents of sexual misconduct (Abbey, 2011). Alcohol consumption and binge-drinking behaviours have consistently been associated with sexual victimisation, with studies linking alcohol use to an increased risk of non-consensual sexual experiences (Howard et al., 2008; Kaysen et al., 2006). Indeed, alcohol has been found to be involved in the majority of hook-up situations (Downing-Matibag & Geisinger, 2009; Lewis et al., 2013; Owen et al., 2010), and may play a role in influencing sexual scripts among college students through the lowering of inhibitions and obscuring judgement in sexual encounters and intimacy (Bersamin et al., 2012; Norman et al., 2014). Alcohol consumption by the victim and/or perpetrator increases the likelihood of sexual violence occurring through its psychological, cognitive and motor effects on an individual (Abbey, 2002). Alcohol consumption may make the victim more vulnerable to experiences of sexual victimisation (Palmer et al., 2010) as perceptions of risk are distorted and adverse interactions may occur as a result of one or both partners being intoxicated.

Socially, intoxication has a complicated and often contradictory relationship in the manner in which it is seen to contribute to sexual violence. For example, victims of sexual assault are often unjustly ascribed responsibility if they were under the influence of alcohol, while responsibility is often diminished on the part of the perpetrator (Abbey & Ortiz, 2008). Societal views on the role of alcohol in sexual circumstances are further confused by the often positive representations of alcohol and sex, and the fact that alcohol is often used by young people to facilitate sexual encounters and produce sexual effects (George & Stoner, 2000). As sexual misconduct often transpires following social interactions involving alcohol or drug use (Bellis et al., 2008; Ullman, 2003), non-consensual sex may occur in scenarios where consensual sex may also have been a possible outcome and may affect a person’s interpretation of the actual situation.

The role of drug use

Like alcohol, drug use can affect decision-making and incur similar consequences (Bellis et al., 2008). Drugs such as cannabis, ecstasy, and cocaine are frequently used in combination with alcohol as part of nightlife socialising, with users using different drugs to achieve a range of outcomes in relation to sexual encounters (Bellis et al., 2008; Calafat et al., 2003). For instance, alcohol is often used to enable sexual encounters, cocaine or ecstasy are used to prolong sex, and cocaine can also be used to enhance the overall sexual experience (Bellis et al., 2008). Cocaine use has been associated with an increased risk of experiencing sexual violence victimisation in a population of women who were perpetrators of interpersonal violence themselves (Stuart et al., 2013). A similar relationship was found in Moore, et al.’s (2008) meta-analysis, which showed a link between women’s, but not men’s, cocaine use and experience of sexual violence victimisation (Moore, et al., 2008). While Moore et al. examined general cocaine use, Stuart et al. examined cocaine use proximal to victimisation (Moore, et al., 2008; Stuart et al., 2013). In the college context less is known about the use and effects of individual drug types in relation to sexual violence as most studies operationalise drug use in general terms and neglect to conduct analyses for specific drug types (Shorey et al., 2011). There is a dearth of research that explores the cumulative effect of mixing alcohol and drugs in the context of sexual violence victimisation (Lawyer et al., 2010). The use of alcohol in conjunction with marijuana, for example, has been highlighted as a specific risk for rape and dating violence victimisation (Messman-Moore et al., 2008; Shorey et al., 2016).

The Irish context

Although a considerable amount of research has been conducted to explore the link between alcohol/drug use and sexual violence, most of this research has been conducted within the US college system (Messman-Moore et al., 2008), and not specifically among college students (Moore, et al., 2008; Stuart et al., 2013). Ireland presents a unique context in which to explore the relationship between substance use and sexual violence. Recreational drug use among college students in Ireland continues to rise and most first-year college students in Ireland report hazardous alcohol consumption (65% of males; 67% of females) (Davoren et al., 2015). The combination of these risk behaviours means that Irish students may be placed at an increased risk of sexual violence as a result.

The current study

While the use of specific drug types has been linked to an increased risk of sexual violence victimisation, particularly in combination with alcohol use, these have not been extensively explored among the college student population in Ireland. The current study explores experiences of sexual violence victimisation among first-year college students, specifically in relation to victim’s reported alcohol and drug consumption. The study will consider:

• Do alcohol and drug use behaviours increase the risk of experiencing different forms of sexual violence?

• Which substance use behaviours incur the greatest risk?

Materials and methods

The campus climate survey methodology developed in the U.S. has provided a major resource for measuring experiences of sexual violence and harassment among college students (Wood et al., 2017), and in recent times multi-campus surveys have been carried out.

Design

Data were collected through the Sexual Experiences Survey, a cross-sectional collaborative study carried out by Active* Consent programme at the University of Galway and the Union of Students’ in Ireland (USI). The questionnaire was based on the Administrator Researcher Campus Climate Collaborative (ARC3) survey tool designed to assess issues of sexual violence and harassment among college students (Swartout et al., 2019). Adaptations were made to the survey to meet the specific needs of Irish college students (terminology, gender-neutral wording, Irish support services and resources). The final questionnaire tool was constructed as an online survey using LimeSurvey software (LimeSurvey GmbH, 2020) and hosted via University of Galway secure servers.

Participants

The final sample of 1778 included in the current analyses were college students, aged between 18 and 25, who were in the first year of third-level education at the time of the survey (Arnett & Tanner, 2006). Table 1 provides a breakdown of the sample according to key demographic variables.

Table 1. Characteristics of first-year undergraduate students, by gender identity.

	Female % (n)	Male % (n)	Non-binary % (n)	Total
Age
18 years	28.3 (339)	26.4 (146)	17.9 (5)	25.6 (490)
19 years	49.2 (589)	47.2 (261)	35.7 (10)	48.4 (860)
20 years	13.0 (156)	16.3 (90)	32.1 (9)	14.3 (255)
21–25 years	9.5 (113)	10.2 (56)	14.4 (4)	9.7 (173)
Sexual orientation
Asexual	5.1 (60)	3.5 (19)	3.6 (1)	4.5 (80)
Bisexual	17.1 (203)	9.5 (52)	39.3 (11)	15.1 (266)
Gay/Lesbian	1.9 (22)	7.9 (43)	14.3 (4)	3.9 (69)
Heterosexual	73.6 (872)	76.2 (417)	0 (0)	73.2 (1289)
Queer/Other	2.4 (28)	2.9 (16)	42.9 (12)	3.2 (56)
Ethnicity
White	93 (1113)	91.7 (507)	82.1 (23)	92.4 (1643)
Black	2 (24)	3.3 (18)	3.6 (1)	2.4 (43)
Asian	2.9 (35)	2.7 (15)	7.1 (2)	2.9 (52)
Other	2.1 (25)	2.4 (13)	7.1 (2)	2.3 (40)
Disability
Yes	15.8 (189)	12.7 (70)	64.3 (18)	15.6 (277)
No	84.2 (1008)	87.3 (483)	35.7 (10)	84.4 (1501)
Relationship status
Single	48.4 (579)	59.0 (326)	53.6 (15)	51.7 (920)
Exclusive relationship	49.7 (595)	38.2 (211)	28.6 (8)	45.8 (814)
Open relationship	1.9 (23)	2.9 (16)	17.9 (5)	2.5 (44)
Total	1197	553	28	1778

Procedure

Twenty-one USI-affiliated higher education institutions (HEIs) in the Republic of Ireland were invited to take part in the study, of which 14 took part. HEI administrators were initially contacted by email providing information about the study and inviting them to take part. Agreeing college administrators then forwarded the study invite email to all students, including all first-year students, containing a link to the online survey. Survey invites were also circulated to students through individual USI channels on each campus and on a national level. Ethical approval for the study was granted by University of Galway Research Ethics Committee (Project identification code: 19-Dec-19).

Measures

Dependent variables

Measures of sexual violence experience in the ARC3 survey were drawn from the Sexual Experiences Survey Short Form Victimisation (SES-SFV) (Koss et al., 2007). Some minor edits were made to the wording of these items to better represent the colloquial language used by young people in Ireland. The final measures can be seen in Table 2.

Table 2. Question wordings for types of sexual violence experienced, and type of tactic used by the perpetrator.

Type of SV^a	Description
Unwanted sexual touching	Someone fondled, kissed, or rubbed up against the private areas of my body (lips, breast/chest, crotch or bottom) or removed some of my clothes without my consent (but did not attempt sexual penetration) by …
Attempted penetration	Even though it didn't happen, someone TRIED to have oral, anal, or vaginal sex with me without my consent by …
Completed penetration (any one of the statements)	Someone had oral sex with me or made me have oral sex with them without my consent by …  Someone put their penis, fingers, or other objects into my vagina without my consent by … ^b Someone put their penis, fingers, or other objects into my anus without my consent by …
Tactic used	Description
Coercion	Telling lies, threatening to end the relationship, threatening to spread rumours about me, making promises I knew were untrue, or continually verbally pressuring me after I said I didn’t want to Showing displeasure, criticising my sexuality or attractiveness, getting angry but not using physical force, after I said I didn’t want to
Incapacitation	Taking advantage of me when I was too drunk or out of it to stop what was happening
Force / threat of force	Threatening to physically harm me or someone close to me Using force, for example holding me down with their body weight, pinning my arms, or having a weapon

^a Response options “0 times”, “1 time”, “2 times” and “3 + times”.

^b Those for whom this question was not relevant were asked to skip this item.

Independent variables

Hazardous alcohol consumption was assessed using the WHO Alcohol Use Disorders Identification Test short form (AUDIT-C) which included three question items (Babor et al., 2001). The first measured frequency of alcohol use; “How often do you have a drink containing alcohol?” with response options “Never; Monthly or less; 2–4 times a month; 2–3 times a week; 4 or more times a week”. Alcohol quantity was measured by asking respondents “How many standard drinks containing alcohol do you have on a typical day when you are drinking?” with the response options “1 or 2; 3 or 4; 5 or 6; 7–9; 10 or more”. Binge drinking was assessed by the item “How often do you have six or more drinks on one occasion?” with response options “Never; Less than monthly; Monthly; Weekly; Daily or almost daily”. Response options on these three items were scored and a total score for alcohol use was calculated ranging from 1 to 12. Different AUDIT-C cut-offs are recommended throughout the literature depending on gender, ethnicity, age and other demographic measures to screen for alcohol use disorders (Reinert & Allen, 2002; Rumpf, 2002). This study used the AUDIT-C tool for the purpose of measuring alcohol consumption in relation to other behaviours only and not as a clinical screening tool, and so the same cut-off was used for the whole study population. A cut-off was applied to total scores to distinguish low-risk alcohol consumption (score < 5), and hazardous alcohol consumption (score of 5 or more) and reflects a common cut-off used in studies among European and college student populations (DeMartini & Carey, 2012; Kelly et al., 2009; Rumpf, 2002; Verhoog et al., 2020).

Drug use was measured by asking respondents “In the last 12 months, have you used any of the following drugs?” with multiple choice response options of “Marijuana or cannabis (grass, pot, hash, weed); Cocaine (Coke, Crack); Ecstasy (E, XTC); Ketamine; I have not used any of these drugs in the last 12 months; Other, please specify”. A binary variable of each drug type was used in the study analysis representing the use of the drug in the last 12 months or not. These items were modified from SLÁN 2007, Survey of Lifestyle, Attitudes & Nutrition in Ireland (Morgan et al., 2008).

Data analysis

Descriptive statistics were calculated to describe the sample demographics (see Table 1), and to describe the prevalence of alcohol use, drug use, and experience of sexual violence behaviours among students (see Tables 3 and 4). Further descriptive analysis assessed the prevalence of type of sexual violence experience by alcohol or drugs use in the last 12 months (see Table 5).

Table 3. Percentage of first-year undergraduate students reporting alcohol and drug use, by gender identity.

	Female % (n)	Male % (n)	Non-binary % (n)
Hazardous alcohol consumption	65.1 (679)	72.2 (342)	47.6 (10)
Cannabis use last 12 months	39.7 (438)	49.6 (256)	50.0 (14)
Cocaine use last 12 months	12.6 (139)	16.3 (84)	3.6 (1)
Ecstasy use last 12 months	9.7 (107)	14.1 (73)	3.6 (1)
Ketamine use last 12 months	5.7 (63)	11.2 (58)	0.0 (0)

Table 4. Prevalence of experience of types of sexual violence, and types of sexual violence tactics, by gender identity.

	Female	Male	Non-binary
SV experience	% (n)	% (n)	% (n)
Unwanted sexual touching	42.4 (507)	22.4 (124)	50.0 (14)
Attempted penetration	25.9 (310)	11.2 (62)	25.0 (7)
Completed penetration	29.4 (352)	9.2 (51)	28.6 (8)
Rape^a	24.6 (294)	8.0 (44)	28.6 (8)
SV tactic experience
Coercion	34.4 (412)	15.7 (87)	46.4 (13)
Incapacitation	35.3 (423)	18.1 (100)	35.7 (10)
Force / threat of force	20.3 (243)	7.6 (42)	25.0 (7)

^a This is a subset of those who reported experience of non-consensual completed penetration.

Table 5. Prevalence of type of sexual violence experience by alcohol or drugs use in the past 12 months, for female and male students.

	% (n)
Sexual violence experience	Sexual violence experience (total)	Hazardous alcohol	Cannabis	Cocaine	Ecstasy	Ketamine
Females	n = 1197	n = 676	n = 438	n = 139	n = 107	n = 63
Unwanted sexual touching	42.4 (507)	51.6 (349)	50.9 (223)	56.8 (79)	59.4 (64)	57.1 (36)
Attempted penetration	25.9 (310)	32.7 (221)	30.4 (133)	36 (50)	36.4 (39)	36.5 (23)
Completed penetration	29.4 (352)	35.7 (241)	35.6 (156)	47.5 (66)	43.9 (47)	47.6 (30)
Rape^a	24.6 (294)	30.8 (208)	30.6 (134)	42.4 (59)	41.1 (44)	41.3 (26)
Males	n = 553	n = 342	n = 256	n = 84	n = 73	n = 58
Unwanted sexual touching	22.4 (124)	26.3 (90)	28.1 (72)	36.9 (31)	34.2 (25)	34.5 (20)
Attempted penetration	11.2 (62)	12.9 (44)	14.5 (37)	22.6 (19)	24.7 (18)	24.1 (14)
Completed penetration	9.2 (51)	11.4 (39)	12.9 (33)	16.7 (14)	20.5 (15)	24.1 (14)
Rape^a	8.0 (44)	10.2 (35)	11.3 (29)	15.5 (13)	19.2 (14)	22.4 (13)

^a This is a subset of those who reported experience of non-consensual completed penetration.

Four separate sets of binary logistic regression models using the “Enter” method were carried out for each of the dependent variables: unwanted sexual touching, non-consensual attempted penetration, non-consensual completed penetration and non-consensual completed penetration specifically through the tactics of incapacitation, force or threat of force. The sub-group of completed penetration through incapacitation, force or threat of force was isolated as it closely aligns with the definition of rape in the Irish legal system (“Criminal Law (Rape) (Amendment) Act, 1990,” 1990). Missing data were included in the sexual violence victimisation measures to avoid over-inflation of the prevalence figures for these sensitive items and to allow for equivalent comparison across the different forms of sexual violence. Missing data on these items ranged from <1% to 11%. Missing data for the substance use items was omitted from analysis and ranged from 7% to 14%.

Regression models were run separately for female (see Table 6) and male students (see Table 7). Due to the relatively small proportion of students that reported the use of harder drugs such as cocaine, ecstasy and ketamine, and the objective of exploring the individual effects of these drugs, separate models were run for each of these three drug types. Due to the existing evidence in the literature linking alcohol use to sexual violence, this variable was included in each model. As the large proportion of participants in the study sample reported hazardous alcohol use, including it as an independent variable allows for the exploration of any additional effect of the other drug variables. The baseline group for the hazardous alcohol use variable consisted of participants who reported no or non-hazardous alcohol consumption, for drug use measures the baseline group consisted of participants reporting no use of the substance being measured. This resulted in a total of 32 regression models. Regression analyses were not carried out for non-binary students due to the small numbers (n = 28). All regression analyses met the assumption of non-multicollinearity and no independent variables were highly correlated, with the collinearity statistics of Tolerance and VIF all falling within accepted limits (Myers, 1990). All statistical analyses were carried out using SPSS version 26 (IBM Corp, 2019).

Table 6. Logistic regression models predicting experience of sexual violence victimisation, for female students.

		Unwanted sexual touching			Attempted penetration			Completed penetration			Rape
		95% CI			95% CI			95% CI			95% CI
	OR	Lower	Upper	OR	Lower	Upper	OR	Lower	Upper	OR	Lower	Upper
Model 1
Alcohol	2.328***	1.769	3.065	2.367***	1.710	3.275	2.069***	1.524	2.810	2.286***	1.638	3.190
Cannabis/marijuana	1.395*	1.079	1.803	1.180	0.889	1.566	1.352*	1.028	1.777	1.367*	1.024	1.823
Model 2
Alcohol	2.351***	1.787	3.092	2.349***	1.697	3.251	1.987***	1.462	2.701	2.183***	1.562	3.050
Cocaine	1.635*	1.078	2.481	1.329	0.904	1.955	2.017***	1.391	2.924	2.089***	1.430	3.050
Model 3
Alcohol	2.376***	1.807	3.124	2.366***	1.711	3.273	2.074***	1.528	2.815	2.256***	1.617	3.148
Ecstasy	1.561*	1.008	2.417	1.330	0.867	2.041	1.631*	1.077	2.470	1.867**	1.225	2.846
Model 4
Alcohol	2.425***	1.846	3.184	2.401***	1.739	3.314	2.105***	1.554	2.852	2.334***	1.676	3.249
Ketamine	1.420	0.841	2.397	1.319	0.769	2.263	1.877*	1.116	3.157	1.817*	1.070	3.085

*p-value <0.05.

**p-value <0.01.

***p-value <0.001.

Table 7. Logistic regression models predicting experience of sexual violence victimisation, for male students.

		Unwanted sexual touching			Attempted penetration			Completed penetration			Rape
		95% CI			95% CI			95% CI			95% CI
	OR	Lower	Upper	OR	Lower	Upper	OR	Lower	Upper	OR	Lower	Upper
Model 1
Alcohol	1.863*	1.070	3.241	1.996	0.900	4.427	1.675	0.748	3.749	1.737	0.739	4.083
Cannabis/marijuana	1.765*	1.121	2.779	1.784	0.958	3.321	2.106*	1.083	4.096	1.955	0.977	3.911
Model 2
Alcohol	1.864*	1.070	3.247	1.858	0.832	4.148	1.730	0.771	3.881	1.741	0.737	4.109
Cocaine	1.953*	1.158	3.296	2.555**	1.340	4.872	1.957	0.979	3.913	2.060*	1.003	4.231
Model 3
Alcohol	1.946*	1.121	3.378	1.831	0.821	4.081	1.648	0.734	3.698	1.655	0.701	3.908
Ecstasy	1.721	0.993	2.982	3.144**	1.639	6.030	2.699**	1.357	5.368	2.859**	1.401	5.836
Model 4
Alcohol	1.960*	1.128	3.407	1.880	0.843	4.195	1.554	0.687	3.514	1.561	0.656	3.717
Ketamine	1.644	0.901	3.002	2.735**	1.352	5.535	3.317**	1.614	6.815	3.460**	1.641	7.297

*p-value <0.05.

**p-value <0.01.

***p-value <0.001.

Results

Alcohol and drug use

Among first-year college students, hazardous alcohol consumption was the most commonly reported substance used in the last 12 months across all gender identities. Cannabis/marijuana use was the most common drug used, and alcohol and drug use rates were higher among male students in all instances.

Experience of sexual violence behaviours

Unwanted sexual touching was the most commonly reported form of sexual violence experienced by first-year students since beginning college. All forms of sexual violence were experienced by a similar proportion of female and non-binary students, with lower prevalence rates among male students in all cases. Sexual violence through incapacitation was the most frequently reported tactic among all students, again notably higher among female and non-binary students.

Prevalence of alcohol use, drug use, and experience of sexual violence

Higher rates of experience of all forms of sexual violence since beginning college were found among female and male students who reported hazardous alcohol consumption, or cannabis/marijuana, cocaine, ecstasy or ketamine use in the last 12 months.

Predictors of sexual violence victimisation for female students

Unwanted sexual touching

All logistic regression models included hazardous alcohol use as an independent variable, this baseline model significantly predicted the experience of unwanted sexual touching for female students, χ²(1) = 45.24, p < .001. When cannabis/marijuana use was added, the model significantly predicted experience of unwanted sexual touching, χ²(2) = 51.67, p < .001. The odds of experiencing unwanted sexual touching were 1.3 times higher for female students who reported cannabis/marijuana use, controlling for the influence of alcohol use. The model containing hazardous alcohol use and cocaine as a predictor was also significant, χ²(2) = 49.27, p < .001. The odds of experiencing unwanted sexual touching were 1.6 times higher for female students who reported cocaine use, controlling for the influence of alcohol use. The model including hazardous alcohol use and ecstasy as predictors was also significant, χ²(2) = 50.69, p < .001. The odds of experiencing unwanted sexual touching were 1.5 times higher for female students who reported ecstasy use, controlling for the influence of alcohol use. When ketamine was added to the hazardous alcohol use model it also significantly predicted experience of unwanted sexual touching, χ²(2) = 49.98, p < .001, however, ketamine did not significantly increase the odds of experiencing unwanted sexual touching. Female students who reported hazardous alcohol consumption were 2.3–2.4 times more likely to experience unwanted sexual touching, depending on the drug being controlled for.

Non-consensual attempted penetration

The baseline model including only hazardous alcohol use as an independent variable significantly predicted the experience of non-consensual attempted penetration for female students, χ²(1) = 32.75, p < .001. None of the drug items incrementally predicted non-consensual attempted penetration over and above hazardous alcohol. Female students who reported hazardous alcohol consumption were 2.3–2.4 times more likely to experience attempted non-consensual penetration, depending on the drug being controlled for.

Non-consensual completed penetration

The baseline model exploring the experience of non-consensual completed penetration among female students including only hazardous alcohol use as a predictor was significant, χ²(1) = 27.91, p < .001. When cannabis/marijuana use was added, the model significantly predicted experience of non-consensual completed penetration, χ²(2) = 32.55, p < .001. The odds of experiencing non-consensual completed penetration were 1.4 times for female students who reported cannabis/marijuana use, controlling for the influence of alcohol use. The model containing hazardous alcohol use and cocaine as a predictor was also significant, χ²(2) = 41.37, p < .001. The odds of experiencing non-consensual completed penetration were 2 times higher for female students who reported cocaine use, controlling for the influence of alcohol use. When ecstasy was added to the hazardous alcohol use model, it was also significant, χ²(2) = 33.14, p < .001. The odds of experiencing non-consensual completed penetration were 1.6 times higher for female students who reported ecstasy use, controlling for the influence of alcohol use. The addition of ketamine to the hazardous alcohol model also significantly predicted experience of non-consensual completed penetration, χ²(2) = 33.43, p < .001. The odds of experiencing non-consensual completed penetration were 1.9 times higher for female students who reported ketamine use, controlling for the influence of alcohol use. Female students who reported hazardous alcohol consumption were 2–2.1 times more likely to experience completed non-consensual penetration, depending on the drug being controlled for.

Rape

The baseline logistic regression model including only hazardous alcohol use as an independent variable significantly predicted the experience of rape for female students, χ²(1) = 30.80, p < .001. When cannabis/marijuana use was added, the model significantly predicted experience of rape, χ²(2) = 35.30, p < .001. The odds of experiencing rape were 1.4 times higher for female students who reported cannabis/marijuana use, controlling for the influence of alcohol use. When cocaine was added to the hazardous alcohol use model, it was also significant, χ²(2) = 44.86, p < .001. The odds of experiencing rape were 2.1 times higher for female students who reported cocaine use, controlling for the influence of alcohol use. The model including hazardous alcohol use and ecstasy as predictors was significant, χ²(2) = 35.63, p < .001. The odds of experiencing rape were 1.9 times higher for female students who reported for ecstasy use, controlling for the influence of alcohol use. The addition of ketamine to the hazardous alcohol model also significantly predicted experience of rape, χ²(2) = 35.50, p < .001. The odds of experiencing rape were 1.8 times higher for female students who reported ketamine use, controlling for the influence of alcohol use. Female students who reported hazardous alcohol consumption were 2.2–2.3 times more likely to experience rape, depending on the drug being controlled for.

Predictors of sexual violence victimisation for male students

Unwanted sexual touching

All logistic regression models included hazardous alcohol use as an independent variable, this baseline model significantly predicted the experience of unwanted sexual touching for male students, χ²(1) = 8.92, p < .01. When cannabis/marijuana use was added, the model significantly predicted experience of unwanted sexual touching, χ²(2) = 14.36, p < .01. The odds of experiencing unwanted sexual touching were 1.8 times higher for male students who reported cannabis/marijuana use, controlling for the influence of alcohol use. The model containing hazardous alcohol use and cocaine as a predictor was also significant, χ²(2) = 14.26, p < .01. The odds of experiencing unwanted sexual touching were 2 times higher for male students who reported cocaine use, controlling for the influence of alcohol use. When ecstasy was added to the hazardous alcohol use model, it was also significant, χ²(2) = 11.80, p < .01. The model including hazardous alcohol use and ketamine also significantly predicted the experience of unwanted sexual touching, χ²(2) = 10.72, p < .001. However, neither ecstasy nor ketamine significantly increased the likelihood of experiencing unwanted sexual touching. Male students who reported hazardous alcohol consumption were 1.8–2 times more likely to experience unwanted sexual touching, depending on the drug being controlled for.

Non-consensual attempted penetration

The baseline model including only hazardous alcohol use as an independent variable significantly predicted the experience of non-consensual attempted penetration for male students, χ²(1) = 5.02, p < .05. When cannabis/marijuana use was added, the model significantly predicted experience of experiencing non-consensual attempted penetration, χ²(2) = 8.49, p < .05, but cannabis/marijuana use did not significantly increase the likelihood of this experience. The model containing hazardous alcohol use and cocaine as predictors was also significant, χ²(2) = 12.60, p < .01. The odds of experiencing non-consensual attempted penetration were 2.6 times higher for male students who reported cocaine use, controlling for the influence of alcohol use. When ecstasy was added to the hazardous alcohol use model, it was also significant, χ²(2) = 15.95, p < .001. The odds of experiencing non-consensual attempted penetration were 3.1 times higher for male students who reported ecstasy use, controlling for the influence of alcohol use. The addition of ketamine to the hazardous alcohol model also significantly predicted experience of non-consensual attempted penetration, χ²(2) = 12.14, p < .01. The odds of experiencing non-consensual attempted penetration were 2.7 times for male students who reported ketamine use, controlling for the influence of alcohol use. Hazardous alcohol use did not increase the likelihood of experiencing non-consensual attempted penetration for male students.

Non-consensual completed penetration

The baseline model exploring the experience of non-consensual completed penetration among male students including only hazardous alcohol use as a predictor was not significant, χ²(1) = , 3.33, p > .05. However, when cannabis/marijuana use was added, the model significantly predicted experience of non-consensual completed penetration, χ²(2) = 8.46, p < .05. The odds of experiencing non-consensual completed penetration were 2.1 times higher for male students who reported cannabis/marijuana use, controlling for the influence of alcohol use. The model containing hazardous alcohol use and cocaine as predictors was also significant, χ²(2) = 6.71, p < .05. Cocaine use did not significantly increase the likelihood of experiencing non-consensual completed penetration. The addition of ecstasy to the hazardous alcohol use model was also significant, χ²(2) = 10.67, p < .01. 2. The odds of experiencing non-consensual completed penetration were seven times higher for male students who reported ecstasy use, controlling for the influence of alcohol use. The model including hazardous alcohol use and ketamine also significantly predicted the experience of non-consensual completed penetration, χ²(2) = 12.94, p < .01. The odds of experiencing non-consensual completed penetration were 3.3 higher for male students who reported ketamine use, controlling for the influence of alcohol use. Hazardous alcohol use did not increase the likelihood of experiencing non-consensual completed penetration for male students.

Rape

The baseline logistic regression model including only hazardous alcohol use as an independent variable did not significantly predict the experience of rape for male students, χ²(1) = 3.2, p > .05. When cannabis/marijuana use was added, the model significantly predicted experience of rape, χ²(2) = 6.95, p < .05, but cannabis/marijuana use did not significantly increase the likelihood of this experience. When cocaine was added to the hazardous alcohol use model, it was also significant, χ²(2) = 6.77, p < .05. The odds of experiencing rape were 2.1 times higher for male students who reported cocaine use, controlling for the influence of alcohol use. The model including hazardous alcohol use and ecstasy as predictors was significant, χ²(2) = 10.75, p < .01. The odds of experiencing rape were 2.9 times higher for male students who reported ecstasy use, controlling for the influence of alcohol use. The addition of ketamine to the hazardous alcohol model also significantly predicted experience of rape, χ²(2) = 12.70, p < .01. The odds of experiencing rape were 3.5 times higher for male students who reported ketamine use, controlling for the influence of alcohol use. Hazardous alcohol use did not increase the likelihood of experiencing rape for male students.

Discussion

This study revealed high prevalence rates of sexual violence victimisation in a sample of first-year college students in the Republic of Ireland. Rates were higher among female and non-binary students, for all forms of sexual violence compared to male students. This supports findings in the U.S. (Cantor et al., 2020), although rates appear to exceed those found in other campus climate studies. However, differences in study design and circumstances may contribute to these variances, such as the measure of sexual violence used, the number of campuses involved, or the recruitment procedure undertaken.

Sexual violence through incapacitation was the most frequently reported tactic overall. This highlights the vulnerabilities faced by students when they are unable to give consent or indeed to withdraw from a sexual situation that they do not wish to take part in. The link between substance use and sexual violence victimisation was further confirmed, with alcohol and drug use being involved in a number of situations.

Hazardous alcohol consumption predicted all forms of sexual violence among female students and increased the likelihood of experiencing unwanted sexual touching for male students. Binge drinking consistently predicts sexual violence victimisation in college (Howard et al., 2008; Kaysen et al., 2006; Thomson Ross et al., 2011), however, it appears to be a particular risk factor for women in the current study. While the reason behind the increased risk for women in this case is unclear, it may be related to biological differences (Erol & Karpyak, 2015) that affect metabolising and blood alcohol concentrations for women. Another factor may be that men are more likely to be perpetrators of sexual violence (Breiding et al., 2015) especially when using alcohol (Abbey et al., 2014).

Cannabis/marijuana predicted experiencing unwanted sexual touching or non-consensual completed penetration through any tactic since beginning college for female and male first-year students. Women were also at an increased risk of experiencing rape, and so the use, and context of use, of these drugs for women may be particularly problematic.

Cocaine use predicted several sexual violence experiences. Cocaine use has previously been related to sexual violence victimisation among women (Moore, et al., 2008; Stuart et al., 2013), but not specifically among the third-level student population. Previous studies failed to provide an explanation for the link between cocaine use and sexual violence victimisation among females, and so further investigation is warranted into the mechanism by which cocaine use places women at increased risk of serious sexual violence. Considerations should include the contexts in which cocaine is used, and if cocaine use patterns differ for women and men, especially considering cocaine use was slightly higher for men in the current study.

Ecstasy or ketamine use in the last 12 months were both significantly associated with an increased likelihood of experiencing non-consensual completed penetration by any tactic, and also specifically through incapacitation, force or threat of force for both female and male students, although the risk for males appears to be greater. This may in part be explained by the higher rates of ecstasy and ketamine use among male students in this study, and further exploration into the why males are more likely to use these drugs may provide important information. While there are few studies linking these drugs and sexual violence victimisation, insights into the contexts that drugs are used could provide a better understanding of why use of these drugs increases the risk of serious sexual assault for men in particular.

While none of the drug types were shown to increase the risk of attempted non-consensual penetration among women, cocaine, ecstasy and ketamine were all associated with an increased likelihood of attempted penetration among men. Given that the overall rates of completed penetration are much higher among female students, this may suggest that male students are less vulnerable to the effects of incapacitation through alcohol or drug use. This may be related to power imbalances or the biological differences in the effects of substances.

This study provides evidence of the increased risk associated with alcohol or drug use and the experience of sexual violence victimisation among first-year college students in Ireland. It highlights the high levels of hazardous alcohol use and some drugs, along with high rates of sexual violence victimisation occurring among college students before they have even completed their first year of college. The findings also highlight how the use of different substances incurs different risks for male and female students. In the existing literature, very little focus is placed on individual drug use among college students and tends to focus on drug use or substance use as a whole. A better understanding of the circumstances and other behaviours associated with different types of drug use among college student groups may provide a better insight into how they interfere with sexual consent communication and positive, safe sexual experiences.

Limitations

This study provides an important insight into the possible effects of alcohol and substance use on sexual violence victimisation, however, it does have a number of limitations. As with all cross-sectional studies, caution must be applied with data in this study and any associations reported do not infer causality. Instead, the study findings should be used to inform and help to prioritise future research in the area of sexual violence victimisation and substance use. Also, due to the relatively small numbers of male students who reported experience of some types of sexual violence and cocaine, ecstasy and ketamine use, it was not possible to include all predictors in one regression model. Instead, separate models had to be carried out for each of these drug types and so it was not possible to explore the cumulative effects of all the drug types. Future studies could benefit from over-sampling of male participants or improving survey promotion among this group as twice as many females completed the current survey, despite the fact that females represent just over half of the third-level population in Ireland (Higher Education Authority, 2021).

Another limitation is that alcohol and drug use measures available in the study are not specific to instances of sexual violence. Future studies including measures of substance use at the time of the event for all forms of sexual violence victimisation would provide more useful and contextual information around the influence of substance use in these particular situations. While the SES-SF does include a question in the follow-up items around victim and perpetrator alcohol or substance use at the time, this is only in reference to the most serious sexual violence experience of the participant. A similar issue arises in the case of the gender of the perpetrator, where it is only attributed in the most serious experience, and so the information is not provided for all forms of sexual violence and is therefore not comparable across experiences.

This study used the AUDIT-C to measure hazardous drinking. This is a commonly used tool for the screening of potential alcohol problems, and while it is useful in this capacity, it is not in itself a diagnostic tool. In the AUDIT measure, binge drinking is described as “6 or more drinks on one occasion” (Babor et al., 2001), however, more recently organisations such as the Centers for Disease Control (CDC) recommend a lower threshold of 5 drinks for men and 4 for women, which would mean that potentially more participants in the current study would have been classified as hazardous drinkers. It must also be noted that the definition of standard drinks in different countries varies, such that the alcohol content in a standard drink in the U.S., where the binge-drinking threshold is 5 drinks, is higher than that of a standard drink in Ireland. And so, while the AUDIT-C questionnaire is a useful screening and exploratory tool, it cannot be considered an accurate measure of hazardous drinking. While it would also be beneficial to explore the current trends on alcohol and drug use in relation to sexual violence victimisation by sexual orientation as well as by gender identity; the current sample does not permit this. Again, the possibility of over-sampling for minority groups may help overcome such issues. This, however, could potentially have repercussions for the overall representativeness of the sample. Another limitation of the current study is that most campus climate surveys in the U.S are often conducted in one particular college using a highly focused recruitment procedure, as opposed to the less targeted multiple-campus approach taken in this study. The emergence of the Covid-19 Pandemic also forced several campuses to withdraw or discontinue their participation in the study. For these reasons, the response rate for the current study is not comparable to those of previous campus climate surveys in the U.S.

Conclusions and recommendations

Victims of sexual violence are in no way responsible for an assault, even if incapacitated in any way through substance use. However, evidence of such links does highlight the importance of substance use education, not only with the aim of reducing the negative effects directly associated with substance use among college students, but also the social effects such as lack of clear consent communication and increased risk of sexual assault. This is even more important considering the changing landscape of socialising which increasingly occurs in private spaces that may be more conducive to the use of drugs and excessive alcohol.

The substance-specific findings highlight the need for a greater understanding of how these substances interact with each other, influence consent communication or promote negative sexual experiences. While there exists an adequate amount of research and information around the role of alcohol in sexual consent, the findings from this study indicate that further investigation would be beneficial in providing a clearer picture around the role of other substances in sexual violence. This should include the context and environments in which different drug types are taken and how this may lead to gender differences in the risks attached to such drug use. The use of a qualitative methodology among users of such drugs would provide the most useful outputs in terms of contributing to consent and substance use education programmes and initiatives.

The findings provide a strong rationale for including alcohol and drug use education as part of future sexual health and consent education programmes for college students. Due to the high rates of both substance use and sexual violence experienced by first-year students, this study emphasises the need to educate young people on these topics in an effective way at the earliest possible time. Ideally, this should occur before young people even start third-level education, but at the very least initiatives should be targeted at incoming first-year students in third-level institutions.

Acknowledgements

We would like to acknowledge the contribution of Róisín O’Donovan (Union of Students in Ireland). We would also like to thank the staff at the third-level institutions who promoted the survey, and most of all the students who took the time to complete the survey.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was funded by the Lifes2good Foundation and the University of Galway.