Skip to main content
SearchLoginLogin or Signup

Development and validation of the MIDSA-SC scale

In recent years, a lot of interest has been devoted to the study of sexual coercion. Although the definition varies across studies, sexual coercion is generally defined as the employment of tactics to obtain sexual activity against freely given consent. Several measures of ...

Published onJan 02, 2024
Development and validation of the MIDSA-SC scale
·

ABSTRACT

In recent years, a lot of interest has been devoted to the study of sexual coercion. Although the definition varies across studies, sexual coercion is generally defined as the employment of tactics to obtain sexual activity against freely given consent. Several measures of sexual coercion exist, however those measures present some methodological limitations. The purpose of this study was to create and validate a scale of sexual coercion using items from the Multidimensional Inventory of Development, Sex, and Aggression and assess its validity. The sample included 529 adult males who were incarcerated for a sexual offence. Results revealed that a 5-item version of the sexual coercion scale had the best psychometric properties, with good internal consistency, convergent and concurrent validity. Furthermore, the item response theory analysis shows that most items were considered difficult and that all items discriminate between individuals at different levels along the continuum.

PRACTICE IMPACT STATEMENT

The construction and validation of a sexual coercion scale would also allow for more appropriate programmes and interventions. A better understanding of the continuum (i.e. to understand the continuum as a continuum of tactics) would help educate individuals about inappropriate sexual behaviours and the concept of consent. To do so, professionals require appropriate instruments to accurately measure this complex problem.

KEYWORDS

Sexual coercion; scale construction; psychometric evaluation; classical test theory; item response theory; validity

Introduction

It is well established that sexual violence is a major social problem that causes serious consequences for the victims, their families, and society (e.g. perpetrators might lie, have difficulty recognising their own aggressive behaviour or wrongly interpret questions; see World Health Organization, 2010). Although both men and women are victims of sexual violence, statistics show that women remain more likely to experience this type of violence (43.6% vs. 24.8%) (Smith et al., 2018). Accord- ing to the survey conducted in 2015 by the National Intimate Partner and Sexual Violence Survey (NISVS), one in five women (21.3%) reports having been a victim of rape or attempted rape in her lifetime in the US (Smith et al., 2018). Still according to the same authors, 43.6% of women report having been victims of some form of non-consensual sexual contact in their lifetime (Smith et al., 2018). Although a large number of people are victims of sexual violence, very few of these incidents are reported to police authorities. As an example, in 2014, only 5% of sexual assaults were reported to police in Canada (Perreault, 2015).

The research community has become increasingly aware that sexual violence is a complex phenomenon and that several forms exist. Sexual violence is not limited to criminal code miscon- duct. In fact, it involves any coercive act of a sexual nature, with or without the use of physical force, directed towards a non-consenting person (Adams-Curtis & Forbes, 2004; Cleveland et al., 1999; Schatzel-Murphy et al., 2009). In recent years, researchers have suggested that the most suit- able term appears to be sexual coercion (Adams-Curtis & Forbes, 2004).

Although the definition largely varies across studies, sexual coercion is generally defined as the employment of tactics to obtain sexual activity against freely given consent. These definitions gen- erally include both verbal and physical tactics. They also include intoxication, such as sexually enga- ging in contact with a person too intoxicated to consent (drugs or alcohol) or giving alcohol or drugs to a person so the person could not object (Adams-Curtis & Forbes, 2004; Camilleri et al., 2009; Cleve- land et al., 1999; Katz et al., 2007; Schatzel-Murphy et al., 2009). A thorough review of the literature on the definition of sexual of sexual coercion, carried out by He and colleagues (2013), identified three key elements of the definition: (1) sexual coercion occurs without the consent of the partner; (2) sexual coercion involves different types of tactics and (3) sexual coercion involves different targeted sexual acts. Several studies on sexual violence have also suggested that certain variables are often, if not always, identified as risk factors for coercive sexual behaviour (Emmers-Sommer & Allen, 1999; Lyndon et al., 2007). The best-known variables are those in the developmental models of sexual aggression proposed by Malamuth and colleagues (1991, 1993, 1995) and Knight and Sims-Knight (2003). According to the confluent model of Malamuth and colleagues, sexual aggression would be the convergence of two pathways: sexual promiscuity and male hostility (Malamuth et al., 1991, 1993, 1995). According to this model, exposure to a violent family environment contributes to, on the one hand, the development of negative schemas towards women and emotional relation- ships and, on the other hand, the development of a hostile masculinity schema accompanied by atti- tudes supportive of violence. Inspired by Malamuth and colleagues, Knight and Sims-Knight’s (2003) model emphasises the role and importance of traits associated with psychopathy and hypersexual- ity. According to this model, exposure to a violent family environment fosters emotional detachment and lack of empathy, on the one hand, and the development of antisocial and aggressive behaviour, on the other. Since sexual aggression and sexual coercion share similar etiological factors, these different variables have also been recognised theoretically relevant to understanding sexual coer- cion (DeGue & DiLillo, 2005).

Academics attempted to conceptualise sexual coercion in order to measure it. Most of them have conceptualised sexual coercion as a continuum, ranging from psychological/verbal pressure to the use of physical force (Adams-Curtis & Forbes, 2004; Christopher, 1988; Farris et al., 2008; Schatzel- Murphy et al., 2009). The term continuum is used because these acts are not a distinct phenomenon and have a certain degree of overlap, which can be viewed along a continuum of sexual violence. There are several ways to study sexual coercion mainly using official records and self-reported measures. Despite the limitations of self-reported measures (e.g. perpetrators might lie, have difficulty recognising their own aggressive behaviour or wrongly interpret questions; see Peterson, 2023), the vast majority of data on sexual coercion comes from self-reported measures because official data do not capture the full range of sexually coercive behaviours. This can be explained by the fact that very few sexual offences are reported to police authorities (Perreault, 2015) and that sexual coercion is not always considered an offence and therefore does not meet the legal criteria of sexual assault (DeGue & DiLillo, 2005).

In the last two decades, many researchers have developed their own measurement instruments, such as the Sexual Coercion Scale (Aalsma et al., 2002), the Postrefusal Sexual Persistence Scale (Struckman-Johnson et al., 2003) and the Sexual Strategies Scale (Peterson et al., 2010). Furthermore, many scales were specifically designed for dating partners and committed relationships, such as the Sexual Coercion in Intimate Relationship Scale (Shackelford & Goetz, 2004), the Tactics to Obtain Sex Scale (Camilleri et al., 2009), and the Multidimensional Sexual Coercion Questionnaire (Raghavan et al., 2015).

To date, two of the most widely used measures are the revised Sexual Experiences Survey (SES) (Koss et al., 1987, 2007; Koss & Oros, 1982) and the Revised Conflict Tactics Scale (CTS2) (Straus, 1990; Straus et al., 1996). These two measures show strong evidence of validity and reliability (Koss et al., 2007; Straus et al., 1996).

Furthermore, the Multidimensional Inventory of Development, Sex, and Aggression (MIDSA, 2011), a contingency-based, computerised inventory, assesses sexual aggression and contains many domains relevant to assess sexual coercion (MIDSA, 2011). These domains include items related to behaviour management problems and impulsive acting out, drug/alcohol use, sexual behaviours from normal to deviant, sexual preoccupation, sexual compulsivity, sexual inadequacy, masculine self-image difficulties, paraphilias, sadism and others. The MIDSA has been widely validated in both clinical and non-clinical samples (Knight & Cerce, 2020; Knight et al., 1994; MIDSA, 2011).

Although these measures are commonly used and potentially useful, they present some meth- odological limitations. First, most were developed in the contexts of casual dating or intimate relationships, thus hardly apply to contexts outside of romantic relationships (Camilleri et al., 2009; Christopher, 1988; Raghavan et al., 2015; Shackelford & Goetz, 2004; Straus et al., 1996). Second, many of those instruments were developed on small samples, which measured mostly United States college students (Christopher, 1988; Mathes & McCoy, 2011; Rapaport & Burkhart, 1984; Waldner et al., 1999). Third, the psychometric properties of these sexual coercion scales have been very little tested and validated (Aalsma et al., 2002; Christopher, 1988; Peterson et al., 2010; Rapaport & Burkhart, 1984; Struckman-Johnson et al., 2003; Waldner et al., 1999). Assessment challenges persist for researchers attempting to evaluate the latent structure and correlates of beha- viours related to sexual coercion. One way to address these challenges is the use of well-established test development techniques on the development and validation. Given the complex character of sexual coercion, we need measurement instruments that have undergone rigorous development and extensive validation testing (DeVellis, 2016).

Sexual coercion is a relatively unknown field of research that raises numerous theoretical and con- ceptual questions. To better understand this complex phenomenon, rigorous development and extensive validation are necessary. Adequate measures are needed for valid research, as “poor measurement imposes an absolute limit on the validity of the conclusions one can reach” (DeVellis, 2016, p. 34). The construction and validation of a sexual coercion scale would also allow for more appropriate programmes and interventions, as well as better assessment capacities. A better under- standing of the continuum (i.e. to understand the continuum as a continuum of tactics) would help educate individuals about inappropriate sexual behaviours and the concept of consent. To do so, professionals require appropriate instruments to accurately measure this complex problem.

Purpose of the study

The purpose of this study was to develop a scale of sexual coercion using items from the MIDSA (MIDSA, 2011) and assess its validity. The MIDSA was selected as it has been widely validated in various populations (Knight & Cerce, 2020; Knight et al., 1994; MIDSA, 2011). Thus, the present study aims to contribute to the literature through: (1) developing a sexual coercion scale from the MIDSA inventory; (2) assessing the psychometric properties of the sexual coercion scale using the classical test theory and the item response theory and (3) establishing its construct and criterion- related validity through associations with theoretically associated scales.

Methods

The database used in this study, including the items used for the construction of the sexual coercion scale, was provided for secondary analyses by psychologist Raymond A. Knight, Professor in the Department of Psychology at Brandeis University. Dr. Knight is the co-creator of the MIDSA, from which the data in this article is derived.

Participants

The sample included 529 adult males who were incarcerated for a sexual offence between 1995 and 2000. The sample was selected from treatment centres where all participants were treated. At the time of the assessment, they were incarcerated in prisons or committed to treatment centres such as the Massachusetts Treatment Center, the Minnesota Correctional Facility-Lino Sex Offender Treatment Program, and the Minnesota Sexual Psychopathic Personality Treatment Center. The sample consists of adult who have sexually offended and does not include juvenile who have sexually offended. The mean age of the sample was 38.98 years (range = 20–68). Their average years of education was 11.0 (SD = 3.0). Fifty percent had never been married. The majority of the sample self-reported as Caucasians (67.3%), followed by African Americans (16.1%), Native Americans (4%), Hispanics (3.8%), and Asians (0.4%). The remaining participants did not report their race (8.4%). The sample included four types of sexual offences: offences against adult victims (35.4%), offences against extrafamilial child victims (29.9%), offences against intrafamilial child victims (24.4%), and offences against mixed age victims (5.3%). Even though some participants (5.1%) lacked sufficient information to be classified, they were also included in the sample. More than half of the participants (64%) had been charged or convicted of a non-sexual offence and 24% had been charged or convicted of a sexual offence as a juvenile. The average age at which they were first incarcerated was 20.2 years (SD = 10.3), the average number of arrests, including the one for the incarceration at the time of testing was 7.9 arrests (SD = 15.1), and the average amount of lifetime incarceration was 4.2 years (SD = 1.8).

Procedure

Participants were verbally solicited by the staff on all treatment centres sites. All interested partici- pants met in groups of 7–12 with research representatives in order to receive more information about the purpose of the study. Participants were informed about the confidentiality of the study and the compensation (US$18) for participating. To ensure that their answers would not be used against them and that their responses would not be shared, participants were told about the Certifi- cate of Confidentiality awarded by the National Institute of Mental Health. Participants were assured that their responses were anonymous and that there was no possibility of linking their research ID to their name. After receiving instructions, participants were invited to complete the computerised version of the MIDSA. This study was approved by Brandeis University and by the three facilities where the participants were tested. Ethical approval from the University of Montreal (#CERSC- 2020-005D) was also obtained.

Sexual coercion

As suggested by Schatzel-Murphy and colleagues (2009), sexual coercion is defined as the employment of tactics aimed to obtain sexual activity against freely given consent. Five broad categories of sexual coercion tactics were assessed: manipulation and bribing, administering alcohol or drugs, taking advantage of someone already intoxicated by drugs or alcohol, using threats of physical force, and use of physical force.

For each of these tactics, four types of outcomes (i.e. sexual activity) were assessed: sex play (fond- ling, kissing or petting), other types of sexual behaviour (oral or anal), attempted sexual intercourse (vaginal or anal), and completed sexual intercourse (vaginal or anal). For each tactic to coerce someone, participants were asked to indicate the frequency of use. Participants rated these items on a Likert-type scale: 0 (Never); 1 (Once); 2 (Sometimes: 2–10 times); 3 (Fairly Often: 11–50 times) and 4 (Very Often: over 50 times). The scale includes a total of 20 items and each subscale (i.e. the 5 sexual coercion tactics) includes 4 items. Possible scores range from 0 to 80 (see Table 1 for items that compose the sexual coercion scale).

Other measures

To demonstrate construct and criterion-related validity, the sexual coercion scale must be correlated with other variables with which the scale should theoretically be associated. The literature suggests that certain variables are often, if not continually, identified as risk factors for sexually coercive behaviour (Emmers-Sommer & Allen, 1999; Lyndon et al., 2007).

For the purposes of this article, the variables generally taken into account in developmental models of sexual aggression proposed by Malamuth and colleagues (1991, 1993, 1995) and Knight and Sims-Knight (2003) were selected in order to validate the coercion scale. These variables were chosen because they are theoretically essential to the understanding of sexual violence. More precisely, variables related to antisocial personality, psychopathy, sexual deviance, and paraphilia were selected. These variables also came from the MIDSA (see Table 1 for items that compose each scale).

The evidence that psychopathy plays a significant role in sexual coercion has continued to grow over the years (Knight & Guay, 2006, 2018). By reviewing the empirical literature on psychopathy and rape, Knight and Guay (2006) indicate that the key symptomatic subdimensions of psychopathy (i.e. lack of perspective taking and empathy, conning and superficial charm and impulsivity) play significant roles in sexually coercive behaviours.

Lack of perspective taking. Lack of perspective taking is a 6-item scale designed to measure the participants’ difficulty to see another’s perspective and to consider both sides of an issue. All items were measured on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal consist- ency of this scale was 0.78. An example of an item is “I find it difficult to see things from the other guy’s point of view”.

Lack of empathy. Lack of empathy is an 8-item scale designed to measure the participants’ lack of feelings of concern for the misfortunes of others. Two items were measured on a 0 (Never) to 4 (Very Often) Likert-type scale while the remaining items were rated on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal consistency of this scale was 0.75. An example of an item is “People who let themselves be conned deserve what they got”.

Conning and superficial charm. Conning and superficial charm is a 6-item scale designed to assess participants conning others, taking advantage of others, manipulating others by lying, and charming others into doing what they want. Two items were measured on a 0 (Never) to 4 (Very Often) Likert-type scale while the remaining items were rated on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal consistency of this scale was 0.79. An example of an item is “I have conned someone to get what I wanted”.

Impulsivity. Impulsivity is a 7-item scale designed to assess participants acting on impulse, losing control, and changing mood. Three items were measured on a 0 (Never) to 4 (Very Often) Likert-type scale while the remaining items were rated on a 0 (Never) to 5 (Very Often) Likert-type scale. The internal consistency of this scale was 0.79. An example of an item is “I have acted impulsively or without thinking”.

Negative masculinity. Negative masculinity is a 5-item scale designed to assess the participants’ attitudes of toughness and masculinity defending their honour. All items were measured on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal consistency of this scale was 0.67. An example of an item is “My friends think of me as being tough”.

Hostility toward women. Hostility toward women is an 8-item scale designed to assess the participants’ negative attitudes towards women and cognitive distortions about rape. All items were measured on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal con- sistency of this scale was 0.88. An example of an item is “Females who get raped probably deserved it”.

Table 1. Items that compose each scale.

A. Sexual coercion scale (20-item scale)

  1. I manipulate or blackmail the other person into doing the following despite lack of consent – sex play.

  2. I manipulate or blackmail the other person into doing the following despite lack of consent – other types of sexual behaviour.

  3. I manipulate or blackmail the other person into doing the following despite lack of consent – attempted sexual intercourse.

  4. I manipulate or blackmail the other person into doing the following despite lack of consent – completed sexual intercourse.

  5. I commit the following acts with an intoxicated person unable to give consent – sex play.

  6. I commit the following acts with an intoxicated person unable to give consent – other types of sexual behaviour.

  7. I commit the following acts with an intoxicated person unable to give consent – attempted sexual intercourse.

  8. I commit the following acts with an intoxicated person unable to give consent – completed sexual intercourse.

  9. I intentionally gave alcohol or drugs to a person so that they are incapable of giving consent for the following acts – sex play.

  10. I intentionally gave alcohol or drugs to a person so that they are incapable of giving consent for the following acts – other sexual behaviour.

  11. I intentionally gave alcohol or drugs to a person so that they are incapable of giving consent for the following acts – attempted sexual intercourse.

  12. I intentionally gave alcohol or drugs to a person so that they are incapable of giving consent for the following acts – completed sexual intercourse.

  13. I threat to use physical force against a person to obtain the following sexual acts – sex play.

  14. I threat to use physical force against a person to obtain the following sexual acts – other types of sexual behaviour.

  15. I threat to use physical force against a person to obtain the following sexual acts – attempted sexual intercourse.

  16. I threat to use physical force against a person to obtain the following sexual acts – completed sexual intercourse.

  17. I use physical force against a person to obtain the following sexual acts – sex play.

  18. I use physical force against a person to obtain the following sexual acts – other sexual behaviour.

  19. I use physical force against a person to obtain the following sexual acts – attempted sexual intercourse.

  20. I use physical force against a person to obtain the following sexual acts – completed sexual intercourse.

B. Psychopathy-related scales

a) Lack of perspective taking (6-item scale)

  1. I try to look at everybody’s side of an argument before I make a decision. (reverse scored)

  2. I believe that every issue has two sides and I try to look at both of them. (reverse scored)

  3. I am always willing to admit when I make a mistake. (reverse scored)

  4. I am quick to admit making a mistake. (reverse scored)

  5. I find it difficult to see things from the “other guy’s” point of view.

  6. No matter who I’m talking to, I’m always a good listener. (reverse scored)

b) Lack of empathy (8-item scale)

  1. It makes me sad to see someone who can’t find anyone to hang out with. (reverse scored)

  2. Seeing someone who is crying makes me feel like crying. (reverse scored)

  3. I have felt very bad about myself after I cheated or did something wrong. (reverse scored)

  4. When I see someone being treated unfairly, I feel sorry for them. (reverse scored)

  5. I have felt sorry after telling people off, even if they deserved it. (reverse scored)

  6. I have had thoughts that made me feel ashamed of myself. (reverse scored)

  7. I feel sorry for people less fortunate than me. (reverse scored)

  8. People who let themselves be conned deserve what they get.

c) Conning and superficial charm (6-item scale)

  1. There have been times when I took advantage of someone.

  2. I have conned someone to get what I wanted.

  3. I have never taken advantage of anyone. (reverse scored)

  4. I have lied to someone to get them to do what I want them to.

  5. I use my charm to get people to notice me.

  6. I can easily charm someone into doing almost anything for me.

d) Impulsivity (7-item scale)

  1. My moods change suddenly.

  2. I have acted impulsively or without thinking.

  3. I do things that make me feel really bad about myself.

  4. I have hurt someone’s feeling by saying something without thinking.

  5. Even though I did not want it, I have lost control of myself.

  6. I have had frightening feelings that I could not understand.

  7. I have gotten in trouble for things that were not my fault.

e) Negative masculinity (5-item scale)

  1. I would beat on a guy who insulted my girl or wife.

  2. My friends think of me as being tough.

  3. I can take a beating as well as any man.

  4. I can hold my own with anybody when it comes to drinking.

  5. I say what’s on my mind, no matter what others may think.

f) Hostility toward women (8-item scale)

  1. Females who get raped probably deserved it.

  2. Girls or women who get drunk at a party are really responsible if someone takes advantage of them sexually.

  3. Girls or women who are raped often had “bad reputations” to begin with.

  4. Because prostitutes sell their bodies for sex anyway, it is not so bad when someone forces them sexually.

  5. If a woman or girl does not strongly resist sexual advances, she is probably willing to have sex.

  6. A real man needs to have sex almost every day.

  7. Most women are cold people.

  8. A man must be boss in a relationship with a woman.

C. Sexualisation related scales

a) Hypersexuality (4-item scale)

  1. I need to masturbate or have sex every day so that I feel less tense.

  2. At times I have almost been driven insane by my thoughts about sex.

  3. There have been times when sex was on my mind so much that I had to make love or masturbate once a day or more.

  4. I sometimes think about sex so much that it gets on my nerves.

b) Anxiety with women (5-item scale)

  1. I feel nervous around females.

  2. It is hard to talk to women or girls.

  3. I feel embarrassed if I talk about sex.

  4. When I have sex with a woman or girl, I feel nervous.

  5. I have had trouble finding someone to have sex with.

c) Sexual compulsivity (9-item scale)

  1. I am not able to control my sexual behaviour.

  2. I have not been able to stop myself from a sexual act, even when I wanted to stop.

  3. I have had a problem controlling my sexual feelings.

  4. I have to fight sexual urges.

  5. Sexual feelings overpower me.

  6. I need to masturbate or have sex every day so that I feel less tense.

  7. I can’t stop thinking about sex.

  8. I am always thinking about sex, no matter where I go or what I do.

  9. I have felt an overpowering urge to do a sexual behaviour that I had thought about.

d) Sexual preoccupation (7-item scale)

  1. While working at a job, my mind will wander to thoughts about sex.

  2. I have sex dreams when I sleep.

  3. When I am bored, I daydream about sex.

  4. Before going to sleep, I think about sex.

  5. I have thought about sex.

  6. There have been times when I thought about sex all of the time.

  7. I get sexually turned on easily.

D. Sexual sadism related scale

a) Agonistic continuum (16-item scale)

  1. Make female do I want turns me on

  2. Thought about forcing sex

  3. Sex thoughts partner tied legs apart

  4. Turns on overpowering sexually

  5. Thought humiliating woman during sex

  6. Sex thoughts threat/fright woman

  7. Good feel hurt someone during sex

  8. More scared more turn on I get

  9. Have tied someone up during sex

  10. Purposely hurt woman physically during sex

  11. During sex enjoy scaring companion

  12. Thought strangling woman during sex

  13. Fantasised do to the person – kill

  14. Thought killing someone during sex

  15. While sex used handcuff/whips/leathers

  16. Beaten woman while having sex

b) Expressive aggression related scales Expressive aggression fantasy (5-item scale)

  1. I have had thoughts about choking a female.

  2. I have thought about threatening or frightening a woman or girl.

  3. When a female rejects me, I get very angry.

  4. When a woman or girl does not do what I want, I get very angry.

  5. Females make me angry.

c) Expressive aggression behaviour (4-item scale)

  1. I have beaten a woman or girl so badly that she had to see a doctor.

  2. I have calmed a woman or girl down with a good slap when she was screaming or crying.

  3. I have roughed up a woman or girl so that she would know that I meant business.

  4. A woman or girl has made me so angry that I have beaten her up.

E. Juvenile delinquency related scales

a) Assaultive behaviour related offences (16-item scale)

  1. Before my 18th birthday, I was involved in physical fights.

  2. Before my 18th birthday, I was involved in physical assault with a man.

  3. Before my 18th birthday, I was involved in physical assault with a woman.

  4. Before my 18th birthday, I was involved in fights while drinking.

  5. Before my 18th birthday, I was involved in verbal abuse while drinking.

  6. Before my 18th birthday, I was involved in assault while drinking.

  7. Before my 18th birthday, I was charged/convicted of a crime while drinking.

  8. Before my 18th birthday, I was involved in assault while high.

  9. Before my 18th birthday, I was charged/convicted of a crime while high.

  10. Before my 18th birthday, I was involved in an event while carried knife.

  11. Before my 18th birthday, I was involved in an event while carried gun.

  12. Before my 18th birthday, I was involved in an event while carried weapon other than gun.

  13. Before my 18th birthday, I used a weapon against someone.

  14. Before my 18th birthday, I was charged/convicted with unarmed robbery.

  15. Before my 18th birthday, I was charged/convicted with armed robbery.

  16. Before my 18th birthday, I was charged/convicted illegal gun possession.

b) Sexual related offences (7-item scale)

  1. Before my 18th birthday, I was charged/convicted with voyeurism.

  2. Before my 18th birthday, I was charged/convicted with any sex offence.

  3. Before my 18th birthday, I was charged/convicted with a sexual offence against a female victim who was a minor.

  4. Before my 18th birthday, I was charged/convicted with a sexual offence against a male victim who was a minor.

  5. Before my 18th birthday, I was charged/convicted with a sexual offence against a female victim who was an adult.

  6. Before my 18th birthday, I was charged/convicted with a sexual offence against a male victim who was an adult.

c ) Drugs/alcohol-related offences (6-item scale)

  1. Before my 18th birthday, I was charged/convicted with any illegal drugs offence related.

  2. Before my 18th birthday, I was charged/convicted with illegal alcohol possession

  3. Before my 18th birthday, I was charged/convicted with illegal drug possession.

  4. Before my 18th birthday, I was charged/convicted with offences involve illegal drug.

  5. Before my 18th birthday, I was charged/convicted with illegal drug using.

  6. Before my 18th birthday, I was charged/convicted with illegal drug trafficking.

F. Adult criminal behaviour related scales

a) Assaultive behaviour related offences (17-item scale)

  1. After my 18th birthday, I was involved in physical fights.

  2. After my 18th birthday, I was involved in physical assault with a man.

  3. After my 18th birthday, I was involved in physical assault with a woman.

  4. After my 18th birthday, I was involved in fights while drinking.

  5. After my 18th birthday, I was involved in verbal abuse while drinking.

  6. After my 18th birthday, I hurt my relative or my partner.

  7. After my 18th birthday, I was involved in assault while drinking.

  8. After my 18th birthday, I was charged/convicted of a crime while drinking.

  9. After my 18th birthday, I was involved in assault while high.

  10. After my 18th birthday, I was charged/convicted of a crime while high.

  11. After my 18th birthday, I was involved in an event while carried knife.

  12. After my 18th birthday, I was involved in an event while carried gun.

  13. After my 18th birthday, I was involved in an event while carried weapon other than gun.

  14. After my 18th birthday, I used a weapon against someone.

  15. After my 18th birthday, I was charged/convicted with unarmed robbery.

  16. After my 18th birthday, I was charged/convicted with armed robbery.

  17. After my 18th birthday, I was charged/convicted illegal gun possession.

b) Sexual related offences (5-item scale)

  1. After my 18th birthday, I was charged/convicted with voyeurism.

  2. After my 18th birthday, I was charged/convicted with any sex offence.

  3. After my 18th birthday, I was charged/convicted with a sexual offence against a female victim who was a minor.

  4. After my 18th birthday, I was charged/convicted with a sexual offence against a male victim who was a minor.

  5. After my 18th birthday, I was charged/convicted with a sexual offence against a female victim who was an adult.

  6. After my 18th birthday, I was charged/convicted with a sexual offence against a male victim who was an adult.

c) Drugs/alcohol-related offences (5-item scale)

  1. After my 18th birthday, I was charged/convicted with illegal alcohol possession

  2. After my 18th birthday, I was charged/convicted with illegal drug possession.

  3. After my 18th birthday, I was charged/convicted with offences involve illegal drug.

  4. After my 18th birthday, I was charged/convicted with illegal drug using.

  5. After my 18th birthday, I was charged/convicted with illegal drug trafficking.

The evidence that sexualisation (i.e. hypersexuality, sexual compulsivity and sexual preoccupation) plays a significant role in sexual coercion has found considerable empirical support (Knight & Sims-Knight, 2003; Malamuth, 1998; Malamuth et al., 1996; Schatzel-Murphy et al., 2009).

Hypersexuality. Hypersexuality is a 4-item scale designed to assess the participants’ sexual drive. All items were measured on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal consistency of this scale was 0.81. An example of an item is “I need to masturbate or have sex every day so that I feel less tense”.

Anxiety with women. Anxiety with women is a 5-item scale designed to measure the partici- pants’ feelings of anxiety, guilt, nervousness, and inadequacy around women and sex. One item was measured on a 0 (Never) to 4 (Very Often) Likert-type scale while the remaining items were rated on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale. The internal consistency of this scale was 0.80. An example of an item is “I feel nervous around females”.

Sexual compulsivity. Sexual compulsivity is a 9-item scale designed to measure participants’ feelings of a lack of control over their sexual behaviours, thoughts, and urges. Three items were measured on a 0 (Definitely False) to 4 (Definitely True) Likert-type scale while the remaining items were rated on a 0 (Never) to 5 (Very Often) Likert-type scale. The internal consistency of this scale was 0.91. An example of an item is “I have to fight sexual urges”.

Sexual preoccupation. Sexual preoccupation is a 7-item scale designed to assess the frequency where participants are thinking, dreaming, and daydreaming about sexual content. All items were scored on a 0 (Never) to 5 (Very Often) Likert-type scale. The internal consistency of this scale was 0.90. An example of an item is “There have been times when I thought about sex all the time”.

The evidence that suggests that sexual sadism and expressive aggression play a significant role in sexual coercion has also found considerable empirical support (Harris et al., 2012; Lalumière et al, 2003; Seto et al., 2012). To date, sexual coercion and sexual sadism are clinically considered as sep- arate constructs, as evidenced by separated diagnoses in the Diagnostic and Statistical Manual of Mental Disorders – 5th Editions (DSM-5; respectively, Paraphilic Coercive Disorder and Sexual Sadism Disorder; American Psychiatric Association [APA], 2013). However, a growing body of litera- ture supports the hypothesis that paraphilic sexual coercion and sexual sadism may in fact represent the same construct (Knight, 2010; Knight et al., 2013; Longpré et al., 2019; Longpré et al., 2020), with sexual coercion at the lower end of the spectrum and sexual sadism at the higher end of the spec- trum. This purported dimension has been labelled the Agonistic Continuum (Knight et al., 2013). The Agonistic Continuum is a single dimension along which individuals can be ordered, from presenting no sexually coercive fantasies or behaviours, to presenting severe forms of sexual sadism (Knight et al., 2013).

Agonistic continuum. Agonistic Continuum is a 16-item scale initially proposed by Knight and colleagues (2013) designed to assess the frequency of thoughts and behaviours ranging from no coercive fantasies or behaviours to nonsadistic sexual coercion to severe sexual sadism.

All items were measured on a Likert-type scale, three items were measured on a 0 (Definitely False) to 4 (Definitely True) scale, one item was measured on a 0 (Never) to 5 (Very Often) scale, while the remaining items were rated on a 0 (Never) to 4 (Very Often) scale. The internal consistency of this scale was 0.91. An example of an item is “When I had sexual thoughts, I thought about threa- tening or frightening a woman or a girl”.

Expressive aggression fantasy. Expressive aggression fantasy is a 5-item scale designed to assess the participants’ frequency of having felt angry toward women and had thoughts of hurting or frigh- tening them in nonsexual situations. All items were scored on a 0 (Never) to 5 (Very Often) Likert-type scale. The internal consistency of this scale was 0.81. An example of an item is “When a female rejects me, I get angry”.

Expressive aggression behaviour. Expressive aggression behaviour is a 4-item scale designed to assess the frequency where participants have beaten or harmed women in nonsexual situations. All items were scored on a 0 (Never) to 5 (Very Often) Likert-type scale. The internal consistency of this scale was 0.80. An example of an item is “I have roughed up a woman or a girl so that she would know that I meant business”.

Some evidence indicates that juvenile delinquency plays a role in predicting the persistence of sexu- ally coercive behaviour into adulthood (Zimring et al., 2007, 2009). In addition, people who commit sexual offences have been often characterised as versatile, meaning that they engage in different types of crimes over their lifetime (Harris, Mazerolle, et al., 2009; Harris, Smallbone, et al., 2009).

Assaultive behaviour related offences. Assaultive behaviour related offences includes 5 scales that measure fighting and assaultive behaviours, robbery charges/convictions, assaultive crime charges/convictions, weapon charges/convictions and illegally carrying weapons. All items were scored on a 0 (Never) to 4 (Very Often) Likert-type scale. This scale includes 16 items, and the internal consistency of this scale was 0.91. An example of an item is “Before my 18th birthday, I was involved in physical fights”.

Sexual related offences. Sexual related offences is a 7-item scale that measures sexual offence charges/convictions with both minors and adults. All items were scored on a 0 (Never) to 4 (Very Often) Likert-type scale. The internal consistency of this scale was 0.80. An example of an item is “Before my 18th birthday, I was charged with a sexual offence against a victim who was a minor”. Drugs/alcohol-related offences. Drugs/alcohol-related offences is a 6-item scale that measures involvement in drugs/alcohol-related offences. All items were scored on a 0 (Never) to 4 (Very Often) Likert-type scale. The internal consistency of this scale was 0.78. An example of an item is “Before my 18th birthday, I was charged with illegal drug trafficking”.

Assaultive behaviour related offences. Assaultive behaviour related offences includes 5 scales that measure fighting and assaultive behaviours, robbery charges/convictions, assaultive crime charges/ convictions, weapon charges/convictions, and carrying weapons. All items were scored on a 0 (Never) to 4 (Very Often) Likert-type scale. This scale includes 17 items and the internal consistency of this scale was 0.90. An example of an item is “After my 18th birthday, I was involved in physical fights”.

Sexual related offences. Sexual related offences is a 5-item scale that measures sexual offence charges/convictions with both minors and adults. All items were scored on a 0 (Never) to 4 (Very Often) Likert-type scale. The internal consistency of this scale was 0.50. An example of an item is “After my 18th birthday, I was charged with a sexual offence against a victim who is a minor”.

Drugs/alcohol-related offences. Drugs/alcohol-related offences is a 5-item scale that measures involvement in drugs/alcohol-related offences. All items were scored on a 0 (Never) to 4 (Very Often) Likert-type scale. The internal consistency of this scale was 0.90. An example of an item is “After my 18th birthday, I was charged with illegal drug trafficking”.

Analysis

In order to develop and validate the sexual coercion scale, three analysis strategies were applied: Clas- sical Test Theory, Item Response Theory, and validity analysis. All statistical analyses were performed with SPSS statistic version 25 (IBM Corp, 2017) and Mplus Version 6.12 (Muthén & Muthén, 19982010).

Classical test theory (CTT)

The classical test theory (CTT) model was used to study the sexual coercion scale’s psychometric properties. CTT describes a set of psychometric procedures used to test items and scale reliability, difficulty, and discrimination (Kline, 2000). CTT assumes that for any trait, every person has a true score. However, because there is no way to observe this true score, the CTT model measures what is called the observed score, which is the sum of the true score and the measurement errors (Kline, 2000). Generally, CTT aims to understand and improve the reliability of psychological tests. To do so, the analysis strategy focuses on the item-to-total correlations for each item and Cronbach’s alpha coefficient for the whole pool of items. Items with non-significant correlations or correlations to the total score below 0.40 should be removed from the scale (Fayers & Machin, 2000). As for Cron- bach’s alpha, the minimum acceptable value for alpha is 0.70, with 0.90 being considered excellent (Kline, 2000). All statistical analyses were performed with SPSS statistic version 25 (IBM Corp, 2017).

Item response theory (IRT)

Item response theory (IRT) methods were used to study the sexual coercion scale’s structural prop- erties. In recent years, IRT became the most important psychometric method to validate scales and it was also presented as a better alternative to CTT (Samejima, 1969). IRT models were used to describe the relationship between the level of the latent trait (i.e. latent construct being measured by the scale), the properties of the items in the scale, and a person’s responses to the individual items in the scale (Reeve & Fayers, 2005). IRT models assume that the examined latent trait, noted as Theta (θ), is unidimensional (i.e. that the items in the scale measure one latent trait) (De Ayala, 2009). In IRT models, a response to an item is influenced by both item characteristics and by the par- ticipant. The relationship between a person’s response to an item and their level on the underlying construct is portrayed by the Item Characteristic Curves (ICC), also called Category Response Curves (Reeve & Fayers, 2005). In this study, we used a 2-Parameter Logistic (2PL) model following Sameji- ma’s Graded Response Model (GRM) (Samejima, 1969, 1997) to assess the sexual coercion scale. The Samejima model is the most commonly used method of estimating IRT models for rating scales with items that use ordered Likert scales (DeVellis, 2016). The 2PL IRT model allows for items to vary in their locations (difficulty parameter) on the latent trait continuum and in their capacity to differen- tiate between persons located at different points on the continuum (discrimination parameter).

The first parameter, beta (b), also called “threshold”, estimates the location of the inflexion point on the ICC, which is the point on the scale where the ICC is steepest. This difficulty parameter, which ranges from −3 to +3, assumes that an item below 0 will be considered easy (or frequently endorsed), while an item above 0 will be considered difficult (or infrequently endorsed) (De Ayala, 2009). The GRM (Samejima, 1969, 1997) generates b estimates for each between-category threshold, which is the level of construct required to pass to the superior level. In other words, the b parameter estimates the point at which the respondent has more chances of endorsing the superior level on the Likert scale. In this study, four thresholds are estimated: (1) Once as opposed to Never; (2) Sometimes as opposed to Once; (3) Fairly Often as opposed to Sometimes and (4) Very Often as opposed to Fairly Often. The second parameter, alpha (α), estimates the discriminant power of the items (Reeve & Fayers, 2005). More precisely, the discriminant parameter indicates how well an item dis- criminates between individuals at different levels along the trait continuum. Discriminant values range from –∞ to +∞, and an item with a good discrimination parameter usually varies from 0.8 to 2.5 (De Ayala, 2009). A discriminant value above 0.5 is also acceptable considering that a value of 0.5 is required to discriminate among respondents (Reeve & Fayers, 2005). 2PL IRT analyses were performed with Mplus Version 6.12 (Muthén & Muthén, 19982010).

Convergent and concurrent validity

In this study, convergent and concurrent validity of the sexual coercion scale were tested using SPSS statistic version 25 (IBM Corp, 2017). Convergent validity is aimed at assessing the degree to which two measures of a construct that theoretically should be related are in fact, related (Cronbach & Meehl, 1955). The convergent validity was evaluated by correlating the sexual coercion scale with the other theoretically associated scales based on Malamuth’s and Knight’s models. A “low” corre- lation (i.e. below 0.30) means that the two scales or components measured non-related constructs, while a “moderate-high” correlation (i.e. between 0.30 and 0.80) means that the sexual coercion measures a construct similar and would suggest a good construct validity (Field, 2009). Concurrent validity refers to the strength of the relationship between a test score and criterion measurements made at the time of test administration (Cronbach & Meehl, 1955). The concurrent validity was eval- uated by correlating the sexual coercion scale with the other measures administered at the same time. The same thresholds as for the convergent validity were used to interpret whether the scale could predict (i.e. between 0.30 and 0.80) or not (i.e. below 0.30) something that was collected approximately at the same time.

Results

Classical test theory

All 20 items of the sexual coercion scale, named MIDSA-Sexual Coercion Scale (MIDSA-SCS), were first analysed in accordance with CTT. The results showed that all items were significantly correlated to the total score, and all items had item-to-total correlations higher than 0.20. The Cronbach’s alpha indicated excellent reliability for this 20-item scale (α = 0.93).

Unidimensionality

The most important assumption of the IRT is unidimensionality, meaning that there is a single trait or dimension underlying performance (De Ayala, 2009). There are several methods to assess unidimen- sionality including exploratory factor analysis. In this study, an exploratory factor analysis was con- ducted using principal components and oblique rotation with the oblimin method. Oblique rotation is privileged since, unlike the orthogonal rotation, it allows for correlation between factors (Tabach- nick & Fidell, 2013). In order to be considered “sufficiently” unidimensional, two indicators are needed: (1) the variance of the first factor needs to be greater than 20% (Reckase, 1979), and (2) the eigenvalue of the first factor must be at least twice as large as the second one (Engelhard, 2013). Analyses conducted on the 20-item version of the scale revealed the presence of 4 factors and that the variance of the first factor was 45.63%, which is greater than the recommended 20%. The eigenvalue for the first factor was 9.13 and the eigenvalue for the second factor was 3.82 which means that the first factor is at least twice as large as the second one. These results indicate that the model is considered “sufficiently” unidimensional and thus, IRT analysis can be performed.

Item response theory

A GRM analysis was then conducted on the 20 initial items of the MIDSA-SCS. The results indicated that a 5-item version (the very short version) represented the best psychometric solution. Fifteen items were deleted because they did not meet the prerequisites for acceptable item parameter esti- mates. Indeed, even if all items have a discriminant value above 0.5 in the 20-item version, 8 items have a value smaller than 0.8, which is the standard for a good discrimination parameter (De Ayala, 2009). By selecting the most discriminating items, we proposed a 5-item with only completed out- comes for each tactic. Other element supported this choice. First, except for administering alcohol or drugs, the completed outcomes for each tactic were the most discriminant. Second, the outcome is not always clear, and the person may not remember what he was after it in the first place. For example, the person may want a complete sexual intercourse but finally stops after oral sex. At this point, if you ask the person what they were after, they will probably say a complete intercourse. Third, the nuance between completed and attempted outcomes are subtle. In other words, it is still difficult to determine what is an attempt and what is a complete act.

Table 2. Item-to-total Correlations and Item Estimates Parameters of the very short version of the MIDSA-SCS.

Item-to-total correlations

p value

Discrimination parameters

Difficulty parameters

 

R

p

α

b1

b2

b3

b4

1. Manipulation and bribing – completed

.78

.001

0.87

−0.64

−0.15

1.10

1.82

2. Taking advantage of someone intoxicated - completed

.73

.001

0.68

0.50

1.19

2.29

3.45

3. Administering alcohol or drugs – completed

.74

.001

0.78

0.94

1.33

2.35

3.12

4. Threats of physical force – completed

.72

.001

2.63

0.46

0.84

1.81

2.42

5. Physical force – completed

.75

.001

2.68

0.33

0.79

1.81

2.25

The MIDSA-SCS showed good reliability (α = 0.80). The results showed that all items were signifi- cantly correlated to the total score. As shown in Table 2, item-to-total correlations ranged from 0.72–0.78. Manipulation and bribingcompleted (r = 0.78) and physical forcecompleted (r = 0.75) were the items that presented the highest item-to-total correlation.

As shown in Table 2, all items have a discrimination value (α) above 0.5 and only 2 items have a value smaller than 0.8. The discrimination parameters ranged from 0.68–2.68, indicating a consider- able variation in item discrimination. Physical force – completed (α = 2.68) was the most discriminat- ing item, indicating that this item does distinguish well among those with lower or higher levels of θ, while taking advantage of someone intoxicated – completed (α = 0.68) was the least discriminating item, indicating that this item does not distinguish well among those with lower or higher levels of θ. The difficulty parameters (b) for the very short version ranged from −0.64 to 3.45 and from −0.64 to 0.94 for the first threshold, from −0.15 to 1.33 for the second, from 1.10 to 2.35 for the third, and from 1.82 to 3.45 for the fourth threshold. The first (the lowest) b value for each item pertains to the prob- ability that a respondent will report engaging in sexually coercive behaviours once, while the last b (the highest) value pertains to the probability that a respondent will report engaging in sexually coercive behaviours very often (i.e. more than 50 times).

Convergent and concurrent validity

Validity analyses consider the relationships between the MIDSA-SCS and various theoretical associ- ated scales identified in previous research (see Table 3). The results showed the expected direction with all significant correlations, except for anxiety with women. The results showed low to moderate- high correlations with related scales, such as psychopathy (r = 0.17–0.41, p < 0.01), sexualisation (r = 0.24–0.31, p < 0.01), juvenile delinquency (r = 0.18–0.32, p < 0.01), and adult delinquency (r = 0.22–0.39, p < 0.01). The results also indicate moderate-high correlations with the agonistic conti- nuum scale (r = 0.46, p < 0.01) and expressive aggression related scales (r = 0.33–0.44, p <0.01).

Discussion

The creation of a sexual coercion scale is consistent with the desire to study broader forms of sexual violence. In recent years, social scientists have attempted to conceptualise it in order to measure it. Although there are various measures of sexual coercion, these measures have several limitations. The goal of this study was to develop and validate a sexual coercion scale and assess its psychometric properties with CTT and IRT analyses.

Psychometric performance of the MIDSA-SCS

The results based on CTT methods show that the internal consistency of the MIDSA-SCS was good as indicated by the Cronbach’s alpha’s coefficient (Fayers & Machin, 2000; Kline, 2000). This result suggests that the instrument coherently investigates sexual coercion as measured by its items. Fur- thermore, all items were correlated positively and significantly to the total score of the scale, indicat- ing that all items are relevant in assessing sexual coercion.

Table 3. Correlations between MIDSA-SCS and other scales.

Long MIDSA-SCS

Short MIDSA-SCS

Very short MIDSA-SCS

R

p

p

P

R

p

Psychopathy related scales

Lack of perspective taking

.32

.001

.27

.001

.22

.001

Lack of empathy

.20

.001

.19

.001

.17

.001

Conning and superficial charm

.44

.001

.40

.001

.37

.001

Impulsivity

.37

.001

.33

.001

.29

.001

Negative masculinity

.21

.001

.25

.001

.25

.001

Hostility toward woman

.45

.001

.42

.001

.41

.001

Sexualisation scales

Hypersexuality

.35

.001

.29

.001

.24

.001

Anxiety with women

.13

.036

.06

.291

.01

.863

Sexual compulsivity

.42

.001

.36

.001

.31

.001

Sexual preoccupation

.33

.001

.29

.001

.24

.001

Sexual sadism scale

Agonistic continuum scale

.54

.001

.50

.001

.46

.001

Expressive aggression scales

Expressive aggression fantasy

.50

.001

.47

.001

.44

.001

Expressive aggression behaviour

.36

.001

.34

.001

.33

.001

Juvenile delinquency scales

Assaultive behaviour related offences

.31

.001

.33

.001

.32

.001

Sexual related offences

.21

.001

.20

.001

.18

.001

Drugs/alcohol-related offences

.28

.001

.29

.001

.28

.001

Adult delinquency scales

Assaultive behaviour related offences

.38

.001

.41

.001

.39

.001

Sexual related offences

.27

.001

.22

.001

.22

.001

Drugs/alcohol-related offences

.28

.001

.27

.001

.26

.001

The results based on the IRT analyses indicate that most of the items included in the MIDSA-SCS were distributed on the positive side of the continuum of difficulty (b > 0), representing more difficult items (or infrequently endorsed). Those results suggest that the MIDSA-SCS mostly assesses the severe end of the sexual coercion continuum. As a reminder, item difficulty refers to the level of the latent construct where an individual has a 50% chance of endorsing a particular response to an item. Consequently, if an item has a higher level of difficulty, it requires a higher level of the latent construct for endorsement. In other words, items that are considered the most difficult are the least frequently endorsed, while items that are considered easier are the most frequently endorsed.

In the very short version, results indicate that the administering alcohol or drugscompleted item emerged as the most difficult item in terms of someone who has tried “Once”, “Sometimes” and “Fairly Often”. As for “Very Often”, the taking advantage of someone intoxicatedcompleted item emerged as the most difficult item. The results suggest that the taking advantage of someone intoxi- catedcompleted and the administrating alcohol or drugscompleted items were on the upper end of the sexual coercion spectrum and were more difficult to attain and less frequent. This result is unexpected considering that intoxication tactics are generally widespread. However, one hypothesis suggests that the studied population may be more inclined to use other tactics of sexual coercion instead of intoxication-related tactics. Convicted samples are represented by people who get caught and the ones that have committed more severe sexual offences. In fact, DeGue and col- leagues (2010) suggest that “sexual aggressors had characteristics that could increase their willing- ness to cross the line and resort to more violent means to obtain sex from an unwilling partner” (p. 402). In addition, as the sample is mostly child molesters (29.9%) and incest offenders (24.4%), another explanation is that intoxication-related tactics might be less used with minor victims rather than adult victims. Finally, the manipulation and bribingcompleted item emerged as the least difficult item for all thresholds. In the very short version, results suggest that items related to manipulation and bribing were at the lower end of the sexual coercion spectrum and were easier to attain and more frequent. This result is consistent with the literature where verbal pressure/ verbal manipulation was identified as the main tactic (Byers & Eno, 1992; Koss et al., 1985; Lyndon et al., 2007; Mosher & Anderson, 1986).

The results based on the IRT analyses also indicate that all items presented proper discriminating power. As a reminder, discrimination parameters provide information about how to distinguish an item between low trait and high-trait individuals. In other words, the discriminant parameter indi- cates how well an item discriminates between individuals at different levels along the trait conti- nuum. The results show that item related to physical force (physical force- completed) had the greatest discrimination parameters, indicating that this item was the most sensitive in distinguishing individual differences in the severity of sexual coercion used. As for the lowest discriminant par- ameters, the taking advantage of someone intoxicatedcompleted item showed the lowest discrimi- nation level and thus, the least sensitivity to individual differences in sexual coercion.

Validity of the MIDSA-SCS

The MIDSA-SCS has demonstrated good convergent and concurrent validity. The MIDSA-SCS shows a high association with variables that have been associated with developmental models of sexual aggression. More precisely, results showed the expected direction with all significant correlations, except for anxiety with women.

The MIDSA-SCS shows low to high levels of correlation with psychopathy related scales, which is consistent with the literature. The evidence that psychopathy plays a significant role in sexual coer- cion has continued to grow over the years (Knight & Guay, 2006, 2018). By reviewing the empirical literature on psychopathy and rape, Knight and Guay (2006) indicate that the key symptomatic sub- dimensions of psychopathy play significant roles in sexually coercive behaviours. To support this, they reviewed evidence from three different domains. First, the literature about general criminal research shows that psychopathic criminals have demonstrated a high risk of sexual coercion. Second, the literature about incarcerated rapists revealed a high incidence of psychopathy. Third, in structural equation models of the etiology of sexual aggression, psychopathy has been identified as traits that define critical paths (Knight & Guay, 2006). Furthermore, many studies have documen- ted associations of psychopathy with sexually coercive acts in different samples, such as convicted and non-convicted samples (Abbey et al., 2011; DeGue et al., 2010; Hersh & Gray-Little, 1998; Jones & Olderbak, 2013; Kosson et al., 1997; Krupp et al., 2012; Mouilso & Calhoun, 2013; Williams et al., 2008). The MIDSA-SCS shows low to high levels of correlation with sexualisation related scales, except for “anxiety with women”. The evidence that sexualisation (i.e. sexual preoccupation, sexual compulsiv- ity, and hypersexuality) plays a significant role in sexual coercion has considerable empirical support (Knight & Sims-Knight, 2003; Malamuth, 1998; Malamuth et al., 1996; Schatzel-Murphy et al., 2009). In fact, various signs of sexualisation have been found to predict sexual coercion among men. For example, in some studies, sociosexuality and sexual dominance emerged as keystones of a male sexual coercion model (Malamuth, 1998; Malamuth et al., 1996; Schatzel-Murphy et al., 2009). Fur- thermore, in other studies, sexual compulsivity proved to be an important predictor as well (Knight & Sims-Knight, 2003; Schatzel-Murphy et al., 2009). Even if we found no significant correlation between sexual coercion and anxiety with women, many studies have found that sexual offenders, mostly child molesters, have social competence deficits such as general social skills, heterosexual and sexual skills, anxiety, and capacity for intimacy (Daversa & Knight, 2007; Emmers-Sommer et al., 2004; Prentky & Knight, 1991). For the purposes of this article, the variables generally con- sidered in the models of Malamuth and colleagues (1991, 1993, 1995) and Knight and Sims- Knight (2003) were selected. As these models were developed to understand sexual offences against women, it would be interesting to add some etiological factors used to explain this type of offences against children and especially on social competence deficits. In this order, the Integrated theory of the etiology of sexual offending by Marshall and Barbaree (1990) and the Pathways Models of Ward and Siegert (2002) could be taken as a reference to measure correlates. The MIDSA-SCS shows moderate-high levels of correlation with expressive aggression related scales and the agon- istic continuum scale, which is also present in the literature. Expressive aggressive behaviours and expressive aggressive fantasies have been found to be significantly higher in sexual offenders than in consensual groups (DeGue et al., 2010; Knight et al., 1985; MIDSA, 2011). Even if sadistic assaults only represent a small proportion of the spectrum of sexual coercion (Knight, 2010, 2014; Knight et al., 2013; Sims-Knight & Guay, 2011), some studies have found that individuals with sadistic interests admit to a higher frequency of sexually coercive behaviour than individuals with no sadistic interests (Harris et al., 2012; Seto et al., 2012). For example, Mokros and colleagues (2014) concluded that individuals who gain sexual gratification by the act of inflicting pain or humiliating others may be predisposed to be sexually aggressive with a nonconsenting partner. In addition, Robertson and Knight (2014) also concluded that sadism is highly correlated with sexual violence. Furthermore, the MIDSA-SCS’s association to the Agonistic Continuum scale tends to add further evidence to the exist- ence of a single construct assessing arousal to coercive or sadistic behaviours, the Agonistic Conti- nuum (Knight et al., 2013). However, it does seem that the MIDSA-SCS may measure more severe sexually coercive behaviours considering its items’ difficulty and discrimination parameters.

Finally, the MIDSA-SCS shows low to high levels of correlation with juvenile delinquency related scales and adult delinquency related scales, which is consistent with the literature. Some evidence indicates that juvenile delinquency plays a role in predicting the persistence of sexually coercive behaviour into adulthood (Parks & Bard, 2006; Zimring et al., 2007, 2009). In addition, people who commit sexual offences have been often characterised as versatile, meaning that they engage in different types of crimes over their lifetime (Harris, Mazerolle, et al., 2009; Harris, Smallbone, et al., 2009).

Limitations

While these results are interesting, several limitations must be acknowledged. First, participants were incarcerated in prisons and special commitment facilities at the time of assessment. In fact, all par- ticipants had been convicted of at least one sexually assaultive crime in their life. However, it should be considered that most measures of sexual coercion were developed on student populations, as presented earlier. As such, the present study is the first to attempt to develop a short measure of sexual coercion based on a convicted sample. While the present findings are based on a convicted sample and may not apply outside of this context, they do bring interesting information regarding sexual coercion in a population that has been excluded from research efforts on this subject. The fact does remain that sexual coercion is a complex phenomenon and because it is not limited to criminal code misconduct, this sample could not be considered a reflection of all sexual coercive persons. The MIDSA-SCS should be validated with different samples, such as general population, student popu- lation, juvenile population, and women population. In order to overcome these limitations, it might be appropriate to recruit a sample from a community-based facility treating individuals who sexually offended or who have a deviant sexual interest (i.e. children) but did not offend. For example, the Intervention centre in sexual delinquency in Quebec (Canada) offers this type of ser- vices for judicialised or non-judicialised individuals.

Second, the MIDSA-SCS should also be validated with other different measures, such as the revised Sexual Experiences Survey (SES) (Koss et al., 1987, 2007; Koss & Oros, 1982) or the Revised Conflict Tactics Scale (CTS2) (Straus, 1990; Straus et al., 1996). The validation with other measures will provide stronger evidence of validity than only with the MIDSA itself.

Third, the MIDSA-SCS does not discriminate between sexual coercion in the context of an intimate relationship or outside of one. In fact, the database does not allow us to consider the context where sexual coercion happened. The database used in this study does not allow us to take into consider- ation if the assaults take place in an intimate context, family context or between strangers. Because the dynamics between romantic partners can be different than between strangers, the MIDSA-SCS should be validated within different sample and different assault context. Also, it could be interest- ing to validate this scale within all types of victims because the tactics used can be different within the types of victims (i.e. children, adults, men, woman). For example, it is possible that some tactics (i.e. intoxication-related tactics) might be less used with minor victims than adult victims.

Four, even if the MIDSA-SCS includes a wider range of tactics, abuse of power or authority (i.e. the use of a position to obtain sexual activity) was not included in this study. The database used in this study does not allow us to take into consideration this type of tactic on its own. Finally, this study includes measures that were based on self-reported measures which come with several limitations. It will be interesting to also access to the person’s official records and compare their criminal charge sheet with their self-reported answers.

Obviously, more research is needed to examine the psychometric properties of the MIDSA-SCS, including repeating the study with multiple samples. Despite these limitations, the MIDSA-SCS pre- sents several strengths. This is the first known study to use CTT and IRT to create and validate a sexual coercion scale. IRT analyses in some cases provided more detailed information than classically based analyses. The construction of the MIDSA-SCS fills a need in the scientific literature, as to date few instruments have been tested via rigorous, scientifically recognised methods. The results, based on the IRT, can help clinicians as well as researchers to identify items in a scale that meaningfully discriminate individuals at different levels of the sexual coercion continuum. In addition, compared to existing measures, the scale of sexual coercion includes only 5 items. The fact that the scale is short means that it could easily be used and easily be applied under different contexts and populations.

Conclusion

In summary, this study has achieved its purpose. Given the complex character of sexual coercion, we need measurement instruments that have undergone rigorous development and extensive vali- dation testing. The goal of this study was to develop and validate a sexual coercion scale that has had rigorous development and extensive validation testing. Based on CTT and IRT, analyses revealed that a 5-item version of the MIDSA-SCS has the best psychometric properties. The MIDSA-SCS encompasses many of the limitations of the existent measures (e.g. limited forms of tactics, context of causal dating, small samples) and has undergone rigorous development and extensive validation testing (e.g. good internal consistency, good discriminating power, good convergent and concurrent validity). More work will be necessary to ensure a complete picture of sexual coer- cion. Future research is needed in order to better understand the conceptual issues of sexual coer- cion. In addition, this scale needs to be validated with other populations and within different contexts.

Disclosure statement

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

References

Aalsma, M. C., Zimet, G. D., Fortenberry, J. D., Blythe, M., & Orr, D. P. (2002). Reports of childhood sexual abuse by ado- lescents and young adults: Stability over time. The Journal of Sex Research, 39(4), 259–263. https://doi.org/10.1080/ 00224490209552149

Abbey, A., Jacques-Tiura, A. J., & LeBreton, J. M. (2011). Risk factors for sexual aggression in young men: An expansion of the confluence model. Aggressive Behavior, 37(5), 450–464. https://doi.org/10.1002/ab.20399

Adams-Curtis, L. E., & Forbes, G. B. (2004). College women’s experiences of sexual coercion: A review of cultural, perpe- trator, victim, and situational variables. Trauma, Violence, & Abuse, 5(2), 91–122. https://doi.org/10.1177/ 1524838003262331

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders DSM-5 (5e éd.). https://doi. org/10.1176/appi.books.9780890425596

Byers, E. S., & Eno, R. J. (1992). Predicting men’s sexual coercion and aggression from attitudes, dating history, and sexual response. Journal of Psychology & Human Sexuality, 4(3), 55–70. https://doi.org/10.1300/J056v04n03_04

Camilleri, J. A., Quinsey, V. L., & Tapscott, J. L. (2009). Assessing the propensity for sexual coaxing and coercion in relationships: Factor structure, reliability, and validity of the tactics to obtain sex scale. Archives of Sexual Behavior, 38(6), 959–973. https://doi.org/10.1007/s10508-008-9377-2

Christopher, F. S. (1988). An initial investigation into a continuum of premarital sexual pressure. Journal of Sex Research, 25(2), 255–266. https://doi.org/10.1080/00224498809551458

Cleveland, H. H., Koss, M. P., & Lyons, J. (1999). Rape tactics from the survivors’ perspective: Contextual dependence and within-event independence. Journal of Interpersonal Violence, 14(5), 532–547. https://doi.org/10.1177/ 088626099014005005

Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281–302. https://doi.org/10.1037/h0040957

Daversa, M. T., & Knight, R. A. (2007). A structural examination of the predictors of sexual coercion against children in adolescent sexual offenders. Criminal Justice and Behavior, 34(10), 1313–1333. https://doi.org/10.1177/ 0093854807302411

De Ayala, R. J. (2009). The theory and practice of item response theory. Guilford Publications. https://books.google.ca/ books?id=-k36zbOBa28C.

DeGue, S., & DiLillo, D. (2005). “You would if you loved me”: Toward an improved conceptual and etiological under- standing of nonphysical male sexual coercion. Aggression and Violent Behavior, 10(4), 513–532. https://doi.org/10. 1016/j.avb.2004.09.001

DeGue, S., DiLillo, D., & Scalora, M. (2010). Are all perpetrators alike? Comparing risk factors for sexual coercion and aggression. Sexual Abuse: A Journal of Research and Treatment, 22(4), 402–426. https://doi.org/10.1177/ 1079063210372140

DeVellis, R. F. (2016). Scale development: Theory and applications. SAGE Publications. https://books.google.ca/books?id= 231ZDwAAQBAJ.

Emmers-Sommer, T. M., & Allen, M. (1999). Variables related to sexual coercion: A path model. Journal of Social and Personal Relationships, 16(5), 659–678. https://doi.org/10.1177/0265407599165006

Emmers-Sommer, T. M., Allen, M., Bourhis, J., Sahlstein, E., Laskowski, K., Falato, W. L., Ackerman, J., Erian, M., Barringer, D., Weiner, J., Corey, J., Krieger, J., Moramba, G., & Cashman, L. (2004). A meta-analysis of the relationship between social skills and sexual offenders. Communication Reports, 6(1), 1–7. doi:10.1080/08934210409389369

Engelhard, G. J. (2013). Invariant measurement. Routledge.

Farris, C., Treat, T. A., Viken, R. J., & McFall, R. M. (2008). Sexual coercion and the misperception of sexual intent. Clinical Psychology Review, 28(1), 48–66. https://doi.org/10.1016/j.cpr.2007.03.002

Fayers, P. M., & Machin, D. (2000). Quality of life: Assessment, analysis and interpretation. John Wiley & Sons Ltd. Field, A. (2009). Discovering statistics using SPSS (3rd ed.). Sage Publications.

Harris, D. A., Mazerolle, P., & Knight, R. A. (2009). Understanding male sexual offending: A comparison of general and specialist theories. Criminal Justice and Behavior, 36(10), 1051–1069. https://doi.org/10.1177/0093854809342242

Harris, D. A., Smallbone, S., Dennison, S., & Knight, R. A. (2009). Specialization and versatility in sexual offenders referred for civil commitment. Journal of Criminal Justice, 37(1), 37–44. https://doi.org/10.1016/j.jcrimjus.2008.12.002

Harris, G. T., Lalumière, M. L., Seto, M. C., Rice, M. E., & Chaplin, T. C. (2012). Explaining the erectile responses of rapists to rape stories: The contributions of sexual activity, non-consent, and violence with injury. Archives of Sexual Behavior, 41(1), 221–229. https://doi.org/10.1007/s10508-012-9940-8

He, S., Tsang, S., & Li, C. (2013). A revision of the sexual coercion in intimate relationships scale for young adults in China. Violence and Victims, 28(3), 483–495. https://doi.org/10.1891/0886-6708.11-00124

Hersh, K., & Gray-Little, B. (1998). Psychopathic traits and attitudes associated with self-reported sexual aggression in college men. Journal of Interpersonal Violence, 13(4), 456–471. https://doi.org/10.1177/088626098013004003

IBM Corp. (2017). IBM SPSS statistics for windows, version 25.0. Jones, D. N., & Olderbak, S. G. (2013). The associations among dark personalities and sexual tactics across different scenarios. Journal of Interpersonal Violence, 29(6), 1050–1070. https://doi.org/10.1177/0886260513506053

Katz, J., Moore, J. A., & Tkachuk, S. (2007). Verbal sexual coercion and perceived victim responsibility: Mediating effects of perceived control. Sex Roles, 57(3-4), 235–247. https://doi.org/10.1007/s11199-007-9253-x Kline, P. (2000). The handbook of psychological testing (2nd ed.). Routledge.

Knight, R. A. (2010). Is a diagnostic category for paraphilic coercive disorder defensible? Archives of Sexual Behavior, 39(2), 419–426. https://doi.org/10.1007/s10508-009-9571-x

Knight, R. A. (2014). Sadism: End of an agonistic continuum [communication orale]. 33rd Conference of the Association for the Treatment of Sexual Abusers (ATSA), San Diego, California.

Knight, R. A., & Cerce, D. D. (2020). Validation and revision of the multidimensional assessment of sex and aggression. Psychologica Belgica, 39(2–3), 135–161. https://doi.org/10.5334/pb.948

Knight, R. A., & Guay, J.-P. (2006). The role of psychopathy in sexual coercion against women. In C. J. Patrick (Ed.), Handbook of psychopathy (pp. 512–532). The Guilford Press.

Knight, R. A., & Guay, J.-P. (2018). The role of psychopathy in sexual coercion against women: An update and expansion. In C. J. Patrick (Ed.), Handbook of psychopathy (pp. 662–681). The Guilford Press.

Knight, R. A., Prentky, R. A., & Cerce, D. D. (1994). The development, reliability, and validity of an inventory for the Multidimensional Assessment of Sex and Aggression. Criminal Justice and Behavior, 21(1), 72–94. https://doi.org/ 10.1177/0093854894021001006

Knight, R. A., Roseberg, R., & Schneider, B. (1985). Classification of sexual offenders – Perspectives, methods, and vali- dation. In A. W. Burgess (Ed.), Research handbook on rape and sexual assault (pp. 222–293). Garland Publishing, Inc. Knight, R. A., Sims-Knight, J., & Guay, J.-P. (2013). Is a separate diagnostic category defensible for paraphilic coercion? Journal of Criminal Justice, 41(2), 90–99. https://doi.org/10.1016/j.jcrimjus.2012.11.002

Knight, R. A., & Sims-Knight, J. E. (2003). The developmental antecedents of sexual coercion against women: Testing alternative hypotheses with structural equation modeling. Annals of the New York Academy of Sciences, 989(1), 72–85. https://doi.org/10.1111/j.1749-6632.2003.tb07294.x

Koss, M. P., Abbey, A., Campbell, R., Cook, S., Norris, J., Testa, M., Ullman, S., West, C., & White, J. (2007). Revising the SES: A collaborative process to improve assessment of sexual aggression and victimization. Psychology of Women Quarterly, 31(4), 357–370. https://doi.org/10.1111/j.1471-6402.2007.00385.x

Koss, M. P., Gidycz, C. A., & Wisniewski, N. (1987). The scope of rape: Incidence and prevalence of sexual aggression and victimization in a national sample of higher education students. Journal of Consulting and Clinical Psychology, 55(2), 162–170. https://doi.org/10.1037/0022-006X.55.2.162

Koss, M. P., Leonard, K. E., Beezley, D. A., & Oros, C. J. (1985). Nonstranger sexual aggression: A discriminant analysis of the psychological characteristics of undetected offenders. Sex Roles, 12(9-10), 981–992. https://doi.org/10.1007/ BF00288099

Koss, M. P., & Oros, C. J. (1982). Sexual experiences survey: A research instrument investigating sexual aggression and vic- timization. Journal of Consulting and Clinical Psychology, 50(3), 455–457. https://doi.org/10.1037/0022-006X.50.3.455

Kosson, D. S., Kelly, J. C., & White, J. W. (1997). Psychopathy-related traits predict self-reported sexual aggression among college men. Journal of Interpersonal Violence, 12(2), 241–254. https://doi.org/10.1177/088626097012002006

Krupp, D. B., Sewall, L. A., Lalumière, M. L., Sheriff, C., & Harris, G. T. (2012). Nepotistic patterns of violent psychopathy: Evidence for adaptation? Frontiers in Psychology, 3, 305–305. https://doi.org/10.3389/fpsyg.2012.00305

Lalumière, M. L., Quinsey, V. L., Harris, G. T., Rice, M. E., & Trautrimas, C. (2003). Are rapists differentially aroused by coercive sex in phallometric assessments? Annals of the New York Academy of Sciences, 989, 211–224. https://doi.org/10. 1111/j.1749-6632.2003.tb07307.x

Longpré, N., Knight, R. A., & Guay, J. P. (2019, July 24–26). Latent structure and covariates of the Agonistic Continuum among community sample and college students [Oral presentation]. 2019 Australian and New Zealand Association for the Treatment of Sexual Abuse (ANZATSA) Conference, Brisbane, Australia.

Longpré, N., Sims-Knight, J., Neumann, C., Guay, J.-P., & Knight, R. A. (2020). Is paraphilic coercion a different construct from sadism or the lower end of an agonistic continuum? Journal of Criminal Justice, 71, Article 101743. https://doi. org/10.1016/j.jcrimjus.2020.101743

Lyndon, A. E., White, J. W., & Kadlec, K. M. (2007). Manipulation and force as sexual coercion tactics: Conceptual and empirical differences. Aggressive Behavior, 33(4), 291–303. https://doi.org/10.1002/ab.20200

Malamuth, N. (1998). An evolutionary-based model integrating research on the characteristics of sexually coercive men. In J. G. Adair, D. Belanger, & K. L. Dion (Eds.), Advances in psychological science (Vol. 1, pp. 151–184). Psychology Press Ltd.

Malamuth, N. M., Heavey, C. L., & Linz, D. (1993). Predicting men’s antisocial behavior against women: The interaction model of sexual aggression. In G. C. N. Hall, R. Hirschman, J. R. Graham, & M. S. Zaragoza (Eds.), Sexual aggression: Issues in etiology, assessment, and treatment (pp. 63–97). Taylor & Francis.

Malamuth, N. M., Heavey, C. L., & Linz, D. (1996). The confluence model of sexual aggression. Journal of Offender Rehabilitation, 23(3–4), 13–37. https://doi.org/10.1300/J076v23n03_03

Malamuth, N. M., Linz, D., Heavey, C. L., Barnes, G., & Acker, M. (1995). Using the confluence model of sexual aggression to predict men’s conflict with women: A 10-year follow-up study. Journal of Personality and Social Psychology, 69(2), 353–369. https://doi.org/10.1037/0022-3514.69.2.353

Malamuth, N. M., Sockloskie, R. J., Koss, M. P., & Tanaka, J. S. (1991). Characteristics of aggressors against women: Testing a model using a national sample of college students. Journal of Consulting and Clinical Psychology, 59(5), 670–681. https://doi.org/10.1037/0022-006X.59.5.670

Marshall, W. L., & Barbaree, H. E. (1990). An integrated theory of the etiology of sexual offending. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 257–275). Plenum Press.

Mathes, E. W., & McCoy, J. (2011). Perpetration of sexual coercion and victim of sexual coercion scales: Development and validation. Psychological Reports, 108(2), 449–469. https://doi.org/10.2466/08.09.16.PR0.108.2.449-469

MIDSA. (2011). Clinical manual. Augur Enterprises, Inc. http://www.midsa.us/pdf/MIDSA_clinical_manual.pdf.

Mokros, A., Schilling, F., Weiss, K., Nitschke, J., & Eher, R. (2014). Sadism in sexual offenders: Evidence for dimensionality. Psychological Assessment, 26(1), 138–147. https://doi.org/10.1037/a0034861

Mosher, D. L., & Anderson, R. D. (1986). Macho personality, sexual aggression, and reactions to guided imagery of rea- listic rape. Journal of Research in Personality, 20(1), 77–94. https://doi.org/10.1016/0092-6566(86)90111-X

Mouilso, E. R., & Calhoun, K. S. (2013). The role of rape myth acceptance and psychopathy in sexual assault perpetration. Journal of Aggression, Maltreatment & Trauma, 22(2), 159–174. https://doi.org/10.1080/10926771.2013.743937

Muthén, L. K., & Muthén, B. O. (1998-2010). Mplus users guide (6th ed.).

Parks, G. A., & Bard, D. E. (2006). Risk factors for adolescent sex offender recidivism: Evaluation of predictive factors and comparison of three groups based upon victim type. Sexual Abuse, 18(4), 319–342. https://doi.org/10.1177/ 107906320601800402

Perreault, S. (2015). La victimisation criminelle au Canada, 2014. https://www150.statcan.gc.ca/n1/fr/pub/85-002-x/ 2015001/article/14241-fra.pdf?st=YEf9aAr6

Peterson, Z. D. (2023). Examining the psychometric properties of the sexual initiation strategies scale (SISS): A new self- report measure of sexual aggression perpetration history. The Journal of Sex Research, 60(1), 91–113. https://doi.org/ 10.1080/00224499.2022.2130856

Peterson, Z. D., Janssen, E., Heiman, J. R., Goodrich, J. T., & Fortenberry, J. D. (2010). Negative affect and emotion dysre- gulation as associated with sexual aggression: A study of young men from urban STD clinics [communication orale]. International Academy of Sex Research, Prague, Czech Republic.

Prentky, R. A., & Knight, R. A. (1991). Identifying critical dimensions for discriminating among rapists. Journal of Consulting and Clinical Psychology, 59(5), 643–661. https://doi.org/10.1037/0022-006X.59.5.643

Raghavan, C., Cohen, S., & Tamborra, T. (2015). Development and preliminary validation of the multidimensional sexual coercion questionnaire (MSCQ). Journal of Sexual Aggression, 21(3), 271–289. https://doi.org/10.1080/13552600.2014. 917341

Rapaport, K., & Burkhart, B. R. (1984). Personality and attitudinal characteristics of sexually coercive college males. Journal of Abnormal Psychology, 93(2), 216–221. https://doi.org/10.1037/0021-843X.93.2.216

Reckase, M. D. (1979). Unifactor latent trait models applied to multifactor tests: Results and implications. Journal of Educational Statistics, 4(3), 207–230. https://doi.org/10.3102/10769986004003207

Reeve, B., & Fayers, P. (2005). Applying item response theory modeling for evaluating questionnaire item and scale prop- erties. In P. Fayers & R. D. Hays (Eds.), Assessing quality of life in clinical trials: Methods and practice (2nd ed., pp. 55–73). Oxford University Press.

Robertson, C. A., & Knight, R. A. (2014). Relating sexual sadism and psychopathy to one another, non-sexual violence, and sexual crime behaviors. Aggressive Behavior, 40(1), 12–23. https://doi.org/10.1002/ab.21505

Samejima, F. (1969). Estimation of latent ability using a response pattern of graded scores. Psychometrika, 34(1), 1–97. https://doi.org/10.1007/BF03372160

Samejima, F. (1997). Graded response model. In W. J. Van der linden & R. K. Hambleton (Eds.), Handbook of item response theory (pp. 85–100). Springer.

Schatzel-Murphy, E. A., Harris, D. A., Knight, R. A., & Milburn, M. A. (2009). Sexual coercion in men and women: Similar beha- viors, different predictors. Archives of Sexual Behavior, 38(6), 974–986. https://doi.org/10.1007/s10508-009-9481-y

Seto, M., Lalumière, M., Harris, G., & Chivers, M. (2012). The sexual responses of sexual sadists. Journal of Abnormal Psychology, 121(3), 739–753. https://doi.org/10.1037/a0028714

Shackelford, T. K., & Goetz, A. T. (2004). Men’s sexual coercion in intimate relationships: Development and initial vali- dation of the sexual coercion in intimate relationships scale. Violence and Victims, 19(5), 541–556. https://doi.org/ 10.1891/vivi.19.5.541.63681

Sims-Knight, J., & Guay, J. P. (2011). Is PCD a construct distinct from sadism? [communication orale]. 30th Annual Conference of the Association for the Treatment of Sexual Abusers (ATSA), Toronto, ON.

Smith, S. G., Zhang, X., Basile, K. C., Merrick, M. T., Wang, J., Kresnow, M., & Chen, J. (2018). National Intimate Partner and Sexual Violence Survey (NIPSV): 2015 Data briefUpdated release. https://www.nsvrc.org/sites/default/files/2021-04/ 2015databrief508.pdf

Straus, M. (1990). The conflict tactics scales and its critics: An evaluation and new data on validity and reliability. In M. A. Straus & R. J. Gelles (Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (pp. 49–73). Transaction Publishers.

Straus, M., Hamby, S., Boney-McCoy, S., & Sugarman, D. (1996). The revised conflict tactics scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17(3), 283–316. https://doi.org/10.1177/ 019251396017003001

Struckman-Johnson, C., Struckman-Johnson, D., & Anderson, P. B. (2003). Tactics of sexual coercion: When men and women won’t take no for an answer. Journal of Sex Research, 40(1), 76–86. https://doi.org/10.1080/00224490309552168

Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics. Pearson.

Waldner, L. K., Vaden-Goad, L., & Sikka, A. (1999). Sexual coercion in India: An exploratory analysis using demographic variables. Archives of Sexual Behavior, 28(6), 523–538. https://doi.org/10.1023/A:1018717216774

Ward, T., & Siegert, R. J. (2002). Toward a comprehensive theory of child sexual abuse: A theory knitting perspective. Psychology, Crime & Law, 8(4), 319–351. https://doi.org/10.1080/10683160208401823

Williams, K. M., Cooper, B. S., Howell, T. M., Yuille, J. C., & Paulhus, D. L. (2008). Inferring sexually deviant behavior from corresponding fantasies. Criminal Justice and Behavior, 36(2), 198–222. ttps://doi.org/10.1177/0093854808327277

World Health Organization. (2010). Violence against women: Intimate partner and sexual violence against women. https://apps.who.int/iris/bitstream/handle/10665/112325/WHO_RHR_14.11_eng.pdf?sequence=1&isAllowed=y.

Zimring, F., Piquero, A., & Jennings, W. (2007). Sexual delinquency in racine: Does early sex offending predict later sex offending in youth and adulthood? Criminology & Public Policy, 6(3), 507-534. https://doi.org/10.1111/j.1745-9133. 2007.00451.x

Zimring, F. E., Jennings, W. G., Piquero, A. R., & Hays, S. (2009). Investigating the continuity of sex offending: Evidence from the second Philadelphia birth cohort. Justice Quarterly, 26(1), 58–76. https://doi.org/10.1080/ 07418820801989734

Comments
0
comment
No comments here
Why not start the discussion?