https://partnerbetrayaltrauma.org/partner-betrayal-trauma-journal-article/

41 min read

Partner Betrayal Trauma Validation Measure & Assess Traumatic Betrayal

The Validation of the Weiss Partner Betrayal Trauma Scale: A Measure to Assess Traumatic Betrayal in Wives of Sex Addicts

Douglas Weiss Ph.D. a and Jamie Stabel b *

• a President, American Association for Sex Addiction Therapy, Colorado Springs, Colorado;

b

Department of Psychology, Capella University, Minneapolis, Minnesota

*720 Elkton Dr. Ste 100, Colorado Springs, CO 80907

719-432-8254

jamiestabel@gmail.com

ABSTRACT

This study was conducted to validate the Weiss Partner Betrayal Trauma Scale (WPBTS); a self-report measure of traumatic spousal betrayal in wives of sex addicts for counselors and researchers. After creating an initial pool of scale items and receiving feedback from a panel of experienced sexual addiction counselors, the WPBTS was edited and pilot tested. Principle component analysis with varimax rotation revealed a simple four-factor structure which accounted for 60.8% of the total variance. Reliability analysis showed sufficient internal consistency (α = .840). Finally, discriminant validity was established using three comparative measures. The WPBTS is established as a measure with strong expert, construct, and discriminant validity. Implications for the use of the WPBTS are discussed.

Keywords: Betrayal, Trauma, Sex Addiction, Infidelity, Intimacy Anorexia

Word Count:7038

The Validation of the Weiss Partner Betrayal Trauma Scale: A Measure to Assess Traumatic Betrayal in Wives of Sex Addicts

Sex Addition (SA) and infidelity significantly impacts the wives of sex addicts (WSA). Research has indicated the revelation of the betrayal of SA or infidelity by a spouse is a traumatic event in the woman's life (Cano & O'Leary, 2000; Laaser, Putney, Bundick, Delmonico, & Griffin, 2017; Özgün, 2010; Steffens & Rennie, 2006). It is estimated that 34 (Özgün, 2010) to 60 percent (Laaser et al., 2017) of women who experience infidelity show significant PTSD symptomology and could be diagnosed with PTSD after a traumatic betrayal such as infidelity. In a study on the traumatic nature of SA disclosure, it was found 69.6% of the sample met the criteria for PTSD diagnosis (Steffens & Rennie, 2006). Ortman (2005) even coined the term Post Infidelity Stress Disorder to describe the constellation of trauma symptoms exhibited by women experiencing relational betrayal or infidelity. However, because trauma is often associated with life-threatening physical violence (i.e. war, physical attack) the psychological trauma of SA and infidelity is often underestimate by spouses, wives, and counselors alike. Yet, trauma is not the only result of SA and infidelity.

Psychological Effects of Betrayal

SA and infidelity disclosure also have a detrimental effect on the mental health of WSA. In one study, not only did infidelity or other humiliating events increase a woman's chances of being diagnosed for major depressive episodes, but WSA also reported a greater frequency of anxiety and depression symptoms than a control group (Cano & O'Leary, 2000). Additionally, high betrayal trauma (e.g. sexual or psychological abuse by a loved one or caregiver) is associated with greater psychological distress than low betrayal trauma (e.g. accident or natural disaster) (Martin et al., 2013). A qualitative study on the causes of depression in women identified betrayal as a primary theme in why women became depressed (Hurst, 1999). Finally, research reported a strong association between betrayal trauma and depression and anxiety symptoms (Tang & Freyd, 2012). Therefore, the trauma of disclosure and these psychological effects are often intertwined, and the more severe the trauma the more severe the psychological effects. However, depression

and anxiety symptoms often persist long after the initial trauma of the disclosure of SA or infidelity.

The depression and anxiety symptoms often persist because traumatic betrayals are a significant and ongoing part of the relationships of WSA and their partners. Unfortunately, research has suggested it is the number of different types of trauma, not the frequency of a repeated trauma which leads to negative outcomes. (Martin et al., 2013). Repeated disclosures, continued sexual acting out, and small everyday breaches of trust continue to traumatize WSA (Steffens & Rennie, 2006). This means, as WSA are confronted with the many different betrayals associated with SA and infidelity, the trauma becomes cumulative and more severe. Therefore, knowing the types and severity of the traumatic betrayal of the WSA is important to understanding their experiences.

Counseling WSA

Traumatic betrayal and the resultant mental distress often drive women to seek counseling (Corely & Schneider, 2002). Research has often identified Sex Addiction (SA) as prevalent and serious problems for marriage and family counselors (Abrahamson et al., 2012; Blow & Hartnett, 2005; Steffens & Rennie, 2006; Whisman et al., 1997). Additionally, infidelity has been recognized as a primary reason couples enter counseling (Peluso & Spina, 2008), yet it is also identified by counselors as one of the most difficult problems to treat (Laaser et al, 2017). Therefore, it is critical for counselors to understand these betrayal traumas and their impact to better assist WSA.While counseling sex addicts and their wives is difficult, literature has provided some guidelines for counselors. For example, Rachman (2010) reminds counselors that it is vital to understand a spouse's current experience of trauma to facilitate treatment. This advice is echoed by Gibson (2008) who also urges counselors to use multiple methods, including formal measures, to assess and understand the betrayal from the aggrieved partner's point of view. Finally, Peluso and Spina (2008) state the number one pitfall for counselors is not all infidelities are alike, so understanding the betrayals is crucial for treatment. These guidelines urge counselors to understand the betrayal from the wives' point of view.

However, for counselors, asking WSA if they have been traumatized by the betrayal does not reveal the true picture of the types of betrayal trauma or the severity of betrayal trauma they have experienced. One woman dealing with sex addiction in her husband may only have to contend with secret masturbation and porn consumption, whereas

another woman is experiencing infidelity, repeated emotional betrayals, a cold, distant relationship, and masturbation. Gibson (2008) recommended using formal measures to assess and understand the betrayed partners experience and perceptions of the betrayal. Further, self-report measures are more accurate than open-ended or interviews in measuring behaviors which carry high social disapproval (Whisman & Wagers, 2005). Yet, there are very few formal measures to assess the severity of betrayal in wives of sex addicts currently available. In fact, a search of the Mental Measurements Yearbook, Google Scholar, and the EBSCO host database found only one measure designed to asses the level of betrayal of a woman in a relationship (Sultan & Muazzam, 2017).

Alternate Measures

The Betrayal Scale for Women was created by Sultan and Muassam (2017) as a 'selfreport measure of intimate partner betrayal for married women' (p. 30). Even though this measure was well validated, there are several problems in adopting this measure for counseling in the United States. First, this measure was created in Pakistan with the Pakistani culture in mind. Further, this measure is 76 items long; quite long for a quick assessment. Finally, the measure is not focused on the betrayals specifically experienced by WSA. It is clear, a new formal measure to assess the severity of betrayal in WSA for counselors of SA and Infidelity is needed. Therefore, because of the traumatic nature of disclosure and repeated betrayals, the severe health outcomes of betrayal, and the need for a measure, the Weiss Partner Betrayal Trauma Scale (WPBTS) is created.

The present study to establish the validity and reliability of the WPBTS was carried out in four phases. In the first phase, an initial item pool was developed and presented to experts to establish expert/face validity. In the second phase, construct validity is explored using factor analysis. In the third phase, reliability statistics are reported. In the fourth phase, evidence of discriminate validity is reported using three measures. Finally, the results and implications are discussed.

Phase I: Initial Development/ Expert Validity

Method

The WPBTS is intended to assess the severity of betrayal trauma in women affected by sexual addiction, intimacy anorexia, and infidelity. First, items were written to assess

sexual addiction betrayals. Although there is no formal definition of SA in the DSM-V, Levine (2010) reminds readers that SA is not a diagnosis but a behavioral complex, and defines sexual addiction as a 'chronic, relapsing disorder in which repeated sexual stimulation persists despite serious negative consequences' (Levine, 2010). While the sex addict 'struggles with sexual behavior in a compulsive or addictive manner' (Weiss D., 2004, p. 61), the manifestation of this struggle is different for each man. However, one constant for these men is the search for sexual stimulation. Therefore, these items included behaviors such as secret porn use or masturbation. An obvious result of the search for sexual stimulation is also infidelity.

Items were written to assess infidelity betrayals. However, rather than just asking if partners were unfaithful, separate items were written for different types of infidelity (i.e. someone known to the spouse, or paid sex) because each of these different types of infidelity has a different underlying emotional issue at its root (Peluso & Spina, 2008). By separating the different types of infidelity, it is hoped the WPBTS will help to identify these issues, and better measure the multiple types of betrayal which may have occurred.

Finally, items were written to assess Intimacy Anorexia betrayals. Intimacy Anorexia is a term coined by Weiss (2010) to describe a specific behavioral phenomenon he was observing in his practice. Weiss defined Intimacy Anorexia as the 'active withholding of emotional, spiritual, and sexual intimacy from the spouse' (p. 12). For some sex addicts, the sexual acting out is an effort to avoid intimacy. For these WSA, additional betrayal traumas include the spouse's avoidance of sexual or emotional intimacy, so items were written to asses these behaviors.

Additional items were also drafted based on betrayal behaviors witnessed by Weiss and colleagues in over 30 years of counseling sex addicts and their partners in these three areas. The WPBTS was then edited. Item language was examined for bias; leading or double-barreled questions were revised; and several unrelated demographic questions were removed.

Participants and Procedure

To begin the validation of the WPBTS, certified sex addiction therapists were contacted to establish expert/face validity. A survey was created on Crowd Signal to evaluate the individual items of the scale with each item listed. Experts rated the validity (not valid,

somewhat valid, and very valid) and provided feedback for improvement of each item. Invitations to validate the scale were sent out to certified sex addiction therapists via LinkedIn. This convivence sampling method yielded N= 38 responses.

Data were collected, cleaned, and entered into SPSS for analysis. Only experts with over 5 years of experience in the field were included in the analysis. Further, surveys with missing values were not included in the analysis. The final analysis included N = 21 experts with an average of over 11 years' experience treating sex-addiction.

Results

Descriptive statistics were created to evaluate the validity of each item. During this phase,

a duplicate item (Item 19) in the survey was identified and eliminated from the analysis. Table 1 includes the item numbers, item text, and frequency of responses by percentage. Overall, the experts judged the measure valid, and all items were judged very valid or somewhat valid by an average of 77% of the experts. However, there were several items with lower perceived validity. [INCLUDE TABLE 1. NEAR HERE]

Items with less than 50% of experts responding 'Very Valid' were identified, and these items are starred in Table 1. Out of the 26 questions in the survey, seven items (Items 3, 10, 14, 15, 23, 24, and 25) fit this criterion and were evaluated. Comments indicated that Items 3, 10, 23, and 24, while somewhat valid, were confusing or vague. Items 3 and 10 were rewritten according to expert comments to increase clarity and objectivity. Items 23 and 24 were also rewritten to make the wording and scoring consistent with the rest of the measures.

One item could not be rewritten. Item 25 was considered Not Valid by almost 43 % of the experts. Comments indicated it was too subjective and too open to interpretation, so it was deleted from the scale.

The final remaining items with lower validity were 14 and 15. These items asked about physical abuse, and experts felt these items measured domestic abuse not betrayal related to SA and infidelity. These items were removed from the scale. With the specific item comments addressed, the more general expert comments were examined.

Several general themes were identified. First, experts indicated the hand, oral, vaginal, anal (HOVA) acronym used in Items 1, 2, and 3 was not widely known and should be simplified. This wording was replaced with the simpler 'sex act of any kind'. Additionally, the 'since with

me' wording in many of the items was considered awkward by experts and eliminated completely. This was resolved by adding an instruction at the beginning of the scale for participants to only consider behaviors which occurred during the relationship. Finally, experts recommended a more parallel construction for several of the questions. Based on this advice, Items 10, 11, and 26 were rewritten to follow the structure of the other items. By creating a more parallel construction in the questions, scoring was also simplified. In the end, most of the items were revised at least a little to simplify, clarify, or create more parallel construction.

After revision, the following items were used in the next phase of pilot testing:

• Item 1. My partner has engaged in sex acts of any kind with a close friend, neighbor, coworker, or acquaintance of mine.
• Item 2. My partner has engaged in sex acts of any kind with someone I do not know.
• Item 3. My partner has paid for live sex services of any kind (e.g. prostitute, stripper).
• Item 4. I have felt manipulated by my partner to participate in sexual behaviors that I was uncomfortable with or regretted.
• Item 5. My partner has viewed porn in secret
• Item 6. My partner has masturbated in secret.
• Item 7. My partner has participated in any type of online sexual relationship.
• Item 8. I believe my partner has a sex addiction of some type.
• Item 9. My partner has had inappropriate emotional relationships with others.
• Item 10. My partner flirts with others in front of me.
• Item 11. When in public, my partner obviously looks at other women.
• Item 12. My partner has been immature emotionally.
• Item 13. My partner has challenges appropriately expressing anger
• Item 14. My partner has been verbally abusive with me.
• Item 15. My partner has avoided emotional intimacy with me.
• Item 16. My partner has avoided sexual intimacy with me.
• Item 17. My partner has been disconnected during sex.
• Item 18. My partner blames me before he takes responsibility for an issue.
• Item 19. My partner has been unduly critical of me.
• Item 20. My partner has avoided sharing his feelings with me.
• Item 21. My partner has avoided apologizing to me when warranted.
• Item 22. My partner makes commitments to me that he does not follow through with.

Phase II. Construct Validity/ Principal Component Analysis

Method

The revised WPBTS was then pilot tested. An invitation to participate in this study was distributed via social media. If interested, participants were directed to the Crowd Signal survey link to provide informed consent and complete the survey. As part of the informed consent the participants were advised of the purpose, requirements, and voluntary nature of participation in the study, advised of the possible (yet minimal) risks, assured of confidentiality and anonymity, and directed to the research author with any questions. Once informed consent was obtained participants completed the survey.

Participants

The recruiting method outlined above resulted in a convenience sample of 154 women, all of whom were impacted by SA. The average age of the sample was 49 years old, but ages ranged from 23 to 74. Most of the respondents were married (79.22%). Only one participant was single, while 9.74 % of the sample were divorced, and 10.39% of the sample were separated. Most of the participants were parents, and only 9.09 % of the women were childless. The greatest percentage of participants (36.36%) reported 2 children, 20.78% reported 3 children, and 26.62% reported 4 or more children. The remaining 7.14% reported only one child. For most of the sample (89.5%), the betrayal was initially discovered over a year ago, and in most of the cases (79.7%), the betrayal was discovered by the wife. While this sample is relatively homogenous (i.e. married women with children who have been betrayed), it is fairly representative of the population the WPBTS is intended to be used on.

Analysis

The overall structure of the WPBTS was explored using principal component analysis (PCA). Examination of the correlation matrix revealed sufficient correlations to justify the use of PCA. Further, the Kaiser-Meyer-Olkin Measure of Sampling Adequacy is .812 or meritorious according to Kaiser (George & Mallery, 2016). The Bartlett test of sphericity is also significant indicating the sample is suitable for PCA (George & Mallery, 2016).

Components were extracted based on a minimum eigenvalue of one, and varimax rotation was applied. Component loadings are displayed in Table 2. [INSERT TABLE 2 NEAR HERE] For ease of interpretation, all loadings below .3 are suppressed. Although a 3factor structure was expected, PCA extracted 5 components in 7 iterations. The 5 components account for 62% of the variation in total betrayal trauma scores. Component 1 could be labeled emotional/Intimacy Anorexia betrayal trauma. Component 2 could be labeled as SA betrayal trauma. Component 3 could be labeled infidelity betrayal trauma. Component 4 only contains two items, and this component has to do with husbands looking or flirting with other women in front of their spouses. One of these items also loads on the emotional betrayal trauma component, and the other also loads on the infidelity betrayal trauma component. Finally, Component 5 could be labeled Intimacy Anorexia betrayal trauma. Several of the items in this component share loadings on Component 1 and Component 2. Because of the theoretical basis of the items, it was wondered if a 3-

factor structure would fit the data better (i.e. SA factor, Intimacy Anorexia factor, Infidelity factor). However, when PCA was run with only 3 factors extracted, no simple factor structure was achieved, and four factors were extracted with an eigenvalue over 1. The results of the analysis suggest a four-factor structure for the WPBTS.

To determine if this proposed four-factor structure is present, PCA is again performed. This time, however, only 4 factors are extracted. Once again, Varimax rotation was employed. Although the 4-component extraction was successful, there were several items which did not load strongly on any factor, and a simple factor structure was not achieved. These items were: Item 11 (looks at other women); Item 4 (manipulated to participate in sexual behaviors); and Item 10 (flirts with other women). Further, the 4 components accounted for only 56.89% of the variance in test scores. This is below the 60% variance threshold set by Hair et al. (2006) for strong measures. Therefore, these items were removed, and the analysis was performed again.

When PCA was performed on the remaining items, the results achieve a simple structure by the criteria outlined by Brown (2009). The rotated structure of the WPBTS is shown in Table 3. Once again, loadings under .3 are suppressed for ease of interpretation. Additionally, items are separated by scales to show which items loaded on which factors. [INSERT TABLE 3. NEAR HERE]

While several of the variables are complex (i.e. load on more than one factor (Brown, 2009)), these overlaps make sense theoretically. For example, many of the Intimacy Anorexia scale items loaded on both the emotional factor and the Intimacy Anorexia factor. This shared loading makes sense theoretically because the nature of these Intimacy Anorexia betrayals is emotional. However, when only 3 factors were extracted, these items did not load on the emotional betrayal component, so the Intimacy Anorexia factor remains separate from the emotional betrayal trauma factor. In the end, the specific nature of the Intimacy Anorexia betrayals separates them from the emotional betrayal factor, but these factors remain related. Because the nature of the measure is to assess the different types of betrayal, the Intimacy Anorexia factor is retained. All items loaded heavily (i.e. > .5) on at least one component.

Results

In the end, PCA reveals a clear structure for the WPBTS. The four scales of the WPBTS

are: The Emotional Betrayal Trauma Scale, the Sexual Addiction Betrayal Trauma Scale, the Infidelity Betrayal Trauma Scale, and the Intimacy Anorexia Betrayal Trauma Scale. Overall, the factor structure accounts for 60.87% of the variance in test scores. This meets the minimum variance accounted for an acceptable factor structure according to Hair et al. (2006). The Emotional Betrayal Trauma Scale accounts for 30.21% of the total variance accounted for by the four-component structure. The Sexual Addiction Betrayal Trauma Scale accounts for 13.87% of the variance, the Infidelity Betrayal Trauma Scale accounts for 10.13% of the variance, and the Intimacy Anorexia Betrayal Trauma Scale accounts for 6.65% of the variance. This is the smallest factor, but, as previously discussed, it is retained to measure this specific type of betrayal trauma. The WPBTS meets the guidelines for a simple factor structure (Brown, 2009) and accounts for an acceptable amount of variance for a social science measure (Hair et al., 2006). With a strong factor structure confirmed, the next step is to establish reliability.

Phase III. Reliability

Method

The internal consistency of the total scale is assessed using reliability analysis in SPSS to calculate Cronbach's α (Cronbach, 1951). The same pilot sample ( N =153) as above is used, but cases with missing items were excluded from the analysis so the valid number was 143. The remaining 19 items (shown in Table 3.) were used to establish the reliability of the WPBTS.

Results

Overall, the WPBTS showed high internal consistency; α= .840. Unsurprisingly, items on the same scale showed higher interitem correlations. Further, deleting any of the items would not increase reliability significantly, and for 18 out of the 19 items, the reliability would be lowered. Therefore, all items are retained, and it is concluded that the scale is reliable and has high internal consistency. Next, the discriminant validity of the WPBTS is explored.

Phase IV. Discriminant validity

In order to establish discriminant validity, several research questions are posed. First, can the WPBTS distinguish the number of betrayals (and thus the severity of betrayal) a woman suffered? It is hypothesized WSA with a higher number of betrayals will have higher WPBTS scores. Next, is the WPBTS associated with marriage satisfaction (CSI) scores? It is hypothesized as WPBTS scores rise; CSI scores will fall. Finally, are higher WPBTS scores associated with PTSD symptoms? It is hypothesized participants with severe clinical PTSD symptoms will have higher WPBTS scores. If the answer to all of these questions is yes, then it can be concluded that the WPBTS measures the severity of betrayal trauma suffered by WSA. Several measures are used to determine the answers to these questions.

Measures Used

In addition to the series of demographic questions, participants also completed the 19 questions WPBTS (shown in Table 3.), the Couples Satisfaction Index (CSI), and Post Traumatic Stress Disorder (PTSD) diagnostic measure. Data were collected, cleaned, and transferred to SPSS. Total WPBTS severity scores were calculated. Next, WPBTS Severity Levels were set. Scores on the WPBTS can range from 0 to 76. Cutoff scores were set by the mean and standard deviation, so scores under 45 (over 1 standard deviation lower than the mean) were set as Substantial Partner Betrayal Trauma Level, scores from 45 to 67 (within one standard deviation of the mean) fell in the Very Substantial Partner Betrayal Trauma Level, scores higher than 67 (over 1 standard deviation above the mean) were set in the Extremely Substantial Partner Betrayal Trauma Level.

Couples Satisfaction Index (CSI)

The marriage satisfaction measure used is the Couples Satisfaction Index (CSI) (Funk & Rogge, 2007). The CSI measures marital satisfaction in 32 questions. The first question assesses the overall happiness of the relationship. The next three questions examine the level of disagreement in the relationship. The next two questions examine how the relationship is going and any regrets the participant may have. Eleven questions are used to evaluate the emotional aspects of the relationship, and four questions are used to evaluate how well the relationship meets the needs of the participant. The next three questions are used to evaluate the quantity and quality of the time spent together. The final seven questions ask the participant to select the word which best describes the

relationship between two words (i.e. interesting/boring or enjoyable/miserable). The items of the CSI were scored according to the directions outlined by the authors which provide an overall happiness score ranging from 0 to 161. The authors state that scores below 104.5 indicate there is significant dissatisfaction in the relationship (Funk & Rogge, 2007).

Post-Traumatic Stress Disorder (PTSD) Assessment Measure

According to the DSM-5, PTSD is diagnosed through 6 different criteria (American Psychiatric Association, 2013). The first criterion is that the person must have experienced 'exposure to actual or threatened death, serious injury, or sexual violence'. In this questionnaire, this criterion is assessed with yes/no item: When I found out about is behaviors, I felt that my marriage/relationship with him was severely threatened or could end. The threatened death of the relationship is the trauma, and this item assesses whether the participants were exposed to this trauma. The trauma is also assessed using a 'check all that apply' item which asked the participants if they experienced intense fear, intense helplessness, or horror. If participants answer yes to the first question and experienced any one of these feelings, then it is concluded that they meet the first criterion for a PTSD diagnosis.

The next criterion is that the person has intrusive distressing memories, distressing dreams, flashbacks, psychological distress from trauma reminders which have persisted for a month or more. This criterion is assessed using 8 items. Four yes/no items address the symptoms, while the remaining four multiple-choice items asses the duration of these symptoms. The presence of intrusive distressing memories is assessed with the item: Since finding out about his behavior, I have had intrusive distressing thoughts about his behavior. The presence of intrusive distressing dreams is assessed through the statement: Since finding out about his behavior, I have had intrusive distressing dreams about his behavior. The presence of flashbacks is assessed through the question: I have experienced reliving the experience of finding out about his behavior or what the behaviors were i.e. flashbacks, illusions, hallucinations like you were actually there again. (This is not just a thought but feeling you are actually there reliving the experience.) The final symptom for this criterion, psychological distress from trauma reminders, is assessed through the statement: I have experienced intense psychological distress when I have externally or internally been exposed to something that symbolizes or resembles his behavior. Finally, the duration of all of these symptoms is addressed through multiple choice questions which asked: If you have experienced (Symptom), it has lasted: (a) days,

(b) weeks, (c) a month, (d) several months, (e) a year, (f) longer than a year. By the DSM5, a person meets the second criterion for a PTSD diagnosis if they experienced two or more of these symptoms for longer than a month after the event.

The third criterion for diagnosing PTSD is that the person engages in persistent behaviors to avoid memories or reminders of the trauma. The longer than a month qualification also continues for this criterion. It is assessed using four items. Two yes/no items address the behaviors, and two multiple-choice items address the duration. The duration multiple choice questions followed the same structure as the previous duration items. The two yes/no items are: I have exerted effort to avoid thoughts, feelings, or conversations about his behavior after disclosure, and I have exerted effort to avoid activities, places or people that arouse recollections of his behavior. If a participant engages in one or both of these behaviors for over a month, then she meets the second criterion for a PTSD diagnosis.

The fourth criterion is 'negative alterations in cognition and mood' marked by symptoms such as diminished interest, emotional restriction, or feelings of detachment. This criterion is assessed using five yes/no items:

• 1. I have an inability to recall aspects of information about his behavior after discovery.
• 2. I have experienced significant diminished interest/participation in important activities after finding out about his behavior.
• 3. After finding out about his betrayal, I have experienced feelings of detachment or estrangement from other people.
• 4. Since finding out about his betrayal, I have felt emotionally restricted.
• 5. Since finding out about his betrayal, I have a sense my future has been shortened.

If the participant answered yes to two or more of these statements, then she met the fourth criterion for a PTSD diagnosis.

The fifth criterion for a PTSD diagnosis is 'marked alterations in arousal and reactivity' marked by behaviors such as irritability, hypervigilance, or sleep disturbances. This is assessed using a 'check all that apply' item, and participants were asked to: Check off all the below that you have experienced since finding out about his behavior with five possible choices: difficulty concentrating, difficulty falling asleep, irritability or outbursts of anger, hypervigilance, or exaggerated startled response. If the participant checked off two or more of these symptoms, then she met the fifth criterion for a PTSD diagnosis.

The final criterion is that the symptoms cause significant distress. This criterion is assessed through one yes/no item: As a result of finding out about his behavior, I have experienced significant distress or impairment in social, occupational or other important areas of functioning in my life. If the participant answered yes, then she met the final criterion for a PTSD diagnosis. While a true diagnosis of PTSD cannot be established through an anonymous online survey, if the participants met all of the criteria for a PTSD diagnosis, it was said that participants were exhibiting clinically severe PTSD symptoms. Together the demographic data, CSI, and PTSD assessment will be used to answer the research questions and establish the discriminant validity of the WPBTS.

Question One

Method

The first question, can the WPBTS distinguish between one betrayal and multiple betrayals, is answered using ANOVA. The Null Hypothesis for the research question is the mean total WPBTS scores will be equal no matter the number of betrayals. The Alternative Hypothesis is there will be a difference in mean WPBTS scores depending on the number of betrayals the woman experienced. An alpha level of .05 is set for this analysis.

Assumptions

The assumptions are tested to ensure the data is suitable for this analysis. The first assumption is that the scores on the quantitative outcome variable (WPBTS scores) are normal. Inspection of the descriptive statistics revealed one outlier which was removed from the analysis. Descriptive statistics were computed again. Both the skew (skew = -.535) and the kurtosis (kurtosis = -.312) are within the less than the absolute value of two value outlined by Warner (2013) for a normal distribution. Additionally, visual inspection of the histogram also shows an approximately normal distribution. Therefore, the data meet the first assumption for ANOVA.

The next two assumptions are that the scores on the outcome variable (WPBTS scores) are independent and participants only belong to one predictor group. Because the WPBTS is not a repeated measure and the data were collected without contact between the participants, the observations are independent. Further, the participants could only

belong to one predictor group (one betrayal or multiple betrayals). Therefore, the data meet the second and third assumption.

The fourth assumption for ANOVA is that the variances are homogenous between groups. This is tested using the Levene test. The results of the Levene Test are not significant ( F (1, 150) = .333, p = .565. The null hypothesis is not rejected, so it is assumed the variances are homogenous, and the data meet the fourth assumption.

The final assumption for ANOVA is that group sizes are equal. The group sizes are not equal in this sample, only 18 participants reported only one betrayal while 134 reported multiple betrayals. Therefore, this assumption is violated. However, while Warner (2013) explains several options for ameliorating this violation, Warner states if the unequal group differences are not caused by an interaction between predictor variables, then using unweighted cell means (i.e. not transforming the data) is a better choice. Because the number of participants in each group is not confounded by some other predictor variable, the group means are accepted unweighted. Therefore, it is concluded that the data are suitable for ANOVA.

Results

A 2 X 1 factorial ANOVA was performed using SPSS to determine if total WPBTS scores differed depending on the number of betrayals a woman reported. Women reporting one betrayal had an average WPBTS score of 46.89, while women reporting multiple betrayals had an average WPBTS score of 58.07. There are obvious differences in the mean WPBTS scores depending on the number of betrayals. Overall, ANOVA results show these differences are significant; F (1, 150) = 17.45, p =.000.

Based on these results, the Alternative Hypothesis is not rejected, and it is concluded there is a significant difference in average WPBTS scores depending on the number of betrayals a woman has suffered. Therefore, the answer to the research question is: Yes, the WPBTS can distinguish the severity of betrayal a woman has suffered, and on average, WSA with higher WPBTS scores also reported more betrayals.

Question Two

Method

The next question posed to establish the discriminant validity of the WPBTS is: Are WPBTS scores associated with marriage satisfaction scores? It is hypothesized that participants with higher WPBTS score will have lower marriage satisfaction scores. This is tested using Pearson's correlation. The Null Hypothesis is: There will be no correlation between WPBTS scores and CSI scores (i.e. r =0).

Assumptions

There are several assumptions for correlation analysis. First, the variables examined must be continuously measured. Both the WPBTS scores and the CSI scores are continuous, so this assumption is met. Next, it is assumed that there are no outliers which may significantly skew the results. The outlier identified previously remained removed from the analysis although analysis with the outlier included was no different. Finally, the scatterplot should show a generally linear relationship with homoscedasticity. Basically, the shape of the scatterplot should be linear, not curved and the scatter of the dots should be in a tubelike shape (not a cone-like) shape (Warner, 2013). The scatterplot comparing the two variables was visually inspected. It was judged the distribution was generally linear and tube-like, so the final assumption was met.

Results

Pearson's Correlational analysis was conducted on the data to determine if WPBTS scores and CSI scores were related. A significant negative correlation was found between WPBTS scores and CSI scores ( r (147) = -.386, p =.000). This means, as hypothesized, as WPBTS scores increased, CSI scores decreased. The Null hypothesis is rejected, and it is concluded that WPBTS scores and CSI scores are negatively correlated. This supports the discriminant validity of the WPBTS. Therefore, the answer to the second question (Are WPBTS scores associated with marriage satisfaction scores?) is: Yes, scores on the WPBTS are negatively correlated with marriage satisfaction scores.

Question Three

Method

This relationship was further examined by dividing the participants into WPBTS severity levels, and then asking if average CSI scores vary significantly based on WPBTS severity

level. ANOVA is conducted to see if there is a difference in average CSI score based on WPBTS severity level. The Null Hypothesis is: There will be no difference in average CSI score based on WPBTS severity level.

Assumptions

The assumptions for ANOVA are discussed above so they will not be detailed again. The data meet the assumptions for ANOVA.

Results

3X1 ANOVA is conducted. The results of the ANOVA show a significant difference in mean CSI score based on WPBTS severity level ( F (2, 144) =6.55, p = .002). Further, post-hoc analysis also showed significant differences between the mean CSI score for the substantial betrayal trauma group ( M= 75.09) and both the very substantial betrayal group ( M= 51.73) and the extremely substantial betrayal group ( M= 41.57). Once again, as WPBTS severity level increases, CSI score decreases. Therefore, the Null hypothesis is rejected, and it is concluded there is a difference in CSI score based on WPBTS severity level. Further, the answer to the research question is: Yes, average CSI scores are different based on WPBTS severity level.

Question Four

Method

The final question asked to establish the discriminant validity of the WPBTS is: Are WPBTS scores associated with PTSD symptoms? Participants were evaluated for PTSD using a self-report measure based on the diagnostic criteria found in the DSM-V detailed above. 65.8% of the sample met the criteria for a PTSD diagnosis and could be said to be exhibiting clinically severe PTSD symptoms. These results are very similar to the study conducted by Steffens and Rennie (2006) who found 69.6% met the criteria for a PTSD diagnosis using a similar measure. ANOVA is performed to see if average WPBTS scores differed based on PTSD diagnosis.

Assumptions

The data meet the assumptions for ANOVA. WPBTS scores are approximately normal, independent, and the groups have homogenous variances.

Results

2X1 ANOVA is conducted. The results of the ANOVA show a significant difference in average WPBTS based on PTSD diagnosis ( F (1, 147) =6.078, p = .015). Although the real-world differences are relatively low (less than 5 points difference), these differences are significant. These results indicate those who could be diagnosed with PTSD have higher WPBTS severity scores than those who could not be diagnosed with PTSD.

Discussion

Overall, the validity and reliability of the WPBTS are well established. The items of the WPBTS were rated valid by experts, and items rated invalid were removed or rewritten. Therefore, the WPBTS shows strong expert/face validity. A simple 4-component structure was revealed by PCA, and 60.87% of the variance in scale scores was accounted for by these 4 components. Therefore, the WPBTS shows strong construct validity. Reliability analysis showed strong internal consistency; α = .840. Therefore, the WPBTS shows strong reliability. Finally, discriminant validity is established. ANOVA shows significant differences in WPBTS scores depending on the number of betrayals. Further, there is a significant negative correlation with marriage satisfaction, and there are significant differences in average marriage satisfaction based on WPBTS severity level. Finally, there is a significant difference in WPBTS scores based on PTSD symptoms severity, and those with a PTSD diagnosis had higher WPBTS severity scores. Therefore, the WPBTS shows strong discriminant validity. Hence, the WPBTS is a valid and reliable measure to assess and examine the types and severity of the betrayal traumas a WSA has experienced.

Implications for Counselors and Researchers

The WPBTS is a valuable tool for both counselors and researchers. For counselors, it provides a snapshot of the trauma experienced by WSA and an additional source of information as suggested by Gibson (2008). It can be used to obtain a betrayal trauma severity score which can, as Rachman (2010) suggested, help counselors to understand 'the person's sense of current serious threat' to facilitate treatment (p. 307). Additionally,

higher WPBTS scores may indicate a need for additional support because clinically severe PTSD symptoms were associated with higher WPBTS scores. The separate infidelity scale can help counselors identify the type of infidelity in the relationship and avoid the most common pitfall for counselors: assuming all infidelities are the same (Peluso & Spina, 2008). Further, the separate Intimacy Anorexia Scale provides information about this 'hidden addiction' which is often overlooked by couples and counselors alike (Weiss, 2016).

For researchers, it provides a valid and reliable measure which can be utilized to study this traumatized population. Further research could determine how WPBTS severity scores impact recovery, sexuality, health. The relationship between the WPBTS and PTSD could be examined in more detail. It would also be illuminating to see how WPBTS scores are associated with treatment outcomes, and if specific treatments are better for extremely substantial betrayal trauma levels. In the end, this research provides strong evidence of the face, construct, and discriminant validity of the WPBTS and a useful instrument for counselors and researchers alike.

Limitations and Suggestions

While this study made significant strides in validating the WPBTS, there are several limitations to this research. First, the sample used in this analysis was relatively homogenous. Almost all of the sample was married, had children, and was mature (average age 49). Given the small homogenous sample, the findings may not be generalizable to other populations. Further, the inclusion criteria only allowed wives of sex addicts to participate. Results may have varied if wives only dealing with only infidelity or Intimacy Anorexia were included. Additional research using larger more varied samples would support these findings.

Next, the WPBTS is a self-report measure and is subject to the biases and limitations associated with these measures (e.g. social desirability bias, reference bias, and etc.). Participants may not have accurately recalled the frequency of behaviors, overreported behaviors out of anger, or underreported behaviors because of ignorance. While the limitations of self-report measures are well known, further research using the WPBTS may reveal these biases, and steps can be taken to mitigate these biases. Although there are

several limitations to this research, this research represents a significant contribution to the evidence supporting the validity of the WPBTS.

References

Abrahamson, I., Hussain, R., Khan, A., & Schofield, M. J. (2012). What helps couples rebuild their relationship after infidelity. Journal of Family Issues, 33 (11), 1494-1519.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: Author.

Blow, A. J., & Hartnett, K. (2005). Infidelity in committed relationships II: A substantive review. Journal of Marital and Family Therapy, 31 (2), 217-233.

Brown, J. D. (2009, Nov.). Choosing the right type of rotation in PCA and EFA. Retrieved from JALT: http://hosted.jalt.org/test/PDF/Brown31.pdf

Cano, A., & O'Leary, K. D. (2000). Infidelity and separations precipitate major depressive episodes and symptoms of nonspecific depression and anxiety. Journal of Consulting and Clinical Psychology, 68 (5), 774-781.

Corely, M. D., & Schneider, J. P. (2002). Disclosing secrets: Guidelines for therapists working with sex addicts and co-addicts. Sexual Addiction and Compulsicity, 9 (1), 43-67.

Cronbach, L. J. (1951). Coefficient Alpha and the internal structure of tests. Psychometrika, 16 , 297-334.

Department of Sociology. (2019, Jan 6). Using the Rosenberg Self-Esteem Scale . Retrieved from University of Maryland: https://socy.umd.edu/using-rosenberg-self-esteemscale

DePrince, A. P., Chu, A. T., & Pineda, A. S. (2011). Links between specific posttrauma appraisals and three forms of trauma-related distress. Psychological Trauma: Theory, Research, Practice, and Policy, 3 (4), 430.

Derbyshire, K. L., & Grant, J. E. (2015). Compulsive sexual behavior: A review of the literature. Journal of Behavioral Addictions, 4 (2), 37-43. doi:10.1556/2006.4.2015.003

Funk, J. L., & Rogge, R. D. (2007). Testing the ruler with item response theory: increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology, 21 (4), 572.

George, D., & Mallery, P. (2016). IBM SPSS statistics 23 step by step: A simple guide and reference (14th ed.). New York, NY: Routledge.

Gibson, D. M. (2008). Relationship betrayal and the influence of religious beliefs: A case illustration of couples counseling. The Family Journal, 16 (4), 344-350.

Gordon, K. C., & Baucom, D. H. (2003). Forgiveness and marriage: Preliminary support for a measure based on a model of recovery from a marital betrayal. American Journal of Family Therapy, 31 (3), 179-199.

Gordon, K. C., Baucom, D. H., & Snyder, D. K. (2004). An integrative intervention for promoting recovery from extramarital affairs. Journal of Marital and Family Therapy, 30 (2), 213-31. doi:10.1111/j.1752-0606.2004.tb01235.x

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate Data Analysis (6th ed.). New York, NY: Pearson Education.

Hurst, S. A. (1999). Legacy of betrayal: A grounded theory of becoming demoralized from the perspective of women who have been depressed. Canadian Psychology, 40 (2), 179191.

Laaser, D., Putney, H. L., Bundick, M., Delmonico, D. L., & Griffin, E. J. (2017). Posttramatic growth in relationally betrayed women. Journal of Marital and Family Therapy, 43 (3), 435-447.

Levine, S. (2010). What is sexual addiction? Journal of Sex & Marital Therapy, 36 (3), 261275.

Martin, C. G., Cromer, L. D., DePrince, A. P., & Freyd, J. J. (2013). The role of cumulative trauma, betrayal, and appraisals in understanding trauma symptomatology. Psychological Trauma: Theory, Research, Practice, and Policy, 5 (2), 110-118.

Ortman, D. C. (2005). Post-infidelity stress disorder. Journal of Psychosocial Nursing & Mental Health Services, 43 (10), 46-54.

Özgün, S. (2010). The predictors of the traumatic effect of extramarital infidelity on married women: Coping strategies, resources, and forgiveness (Doctoral Dissertation, Middle East Technical University). Retrieved from https://www.researchgate.net

Peluso, P. R., & Spina, P. (2008). Understanding infidelity: Pitfalls and lessons for couples counselors. The Family Journal, 16 (4), 324-327. doi: https://doi.org/10.1177/1066480708323282

Rachman, S. (2010). Betrayal: A psychological analysis. Behaviour Research and Therapy, 48 (4), 304-311. doi:10.1016/j.brat.2009.12.002

Spitzer, R. L., Williams, J. B., & Kroenke, K. (2019). PHQ-SADS. Retrieved from PHQ Screeners: https://www.phqscreeners.com

Steffens, B. A., & Rennie, R. L. (2006). The traumatic nature of disclosure for wives of sexual addicts. Sexual Addiction & Compulsivity, 13 (2/3), 247-267.

Sultan, H., & Muazzam, A. (2017). Development and validation of betrayal scale for women. Journal of Behavioral Sciences, 27 (2), 30-47.

Sussman, S., Lisha, N., & Griffiths, M. (2011). Prevalence of the addictions: A problem of the majority or the minority? Evaluation & the Health Professions, 34 (1), 3-56.

Tang, S., & Freyd, J. J. (2012). Betrayal trauma and gender differences in posttraumatic stress. Psychological Trauma: Theory, Research, Practice, and Policy, 4 (5), 469-478.

Warner, R. M. (2013). Applied statistics: From bivariate through multivariate techniques (2nd ed.). Thousand Oaks, CA: Sage. doi:ISBN: 9781412991346

Weiss, D. (2004). The prevalence of depression in male sex addicts residing in the United States. Sexual Addiction & Compulsivity, 11 (1/2), 57-69.

Weiss, D. (2016). Intimacy anorexia: Healing the hidden addiction in your marriage. Colorado Springs, Colorado: Discovery Press.

Whisman, M. A., & Wagers, T. P. (2005). Assessing relationship betrayals. Journal of Clinical Psychology, 61 (11), 1383-1391.

Whisman, M. A., Dixon, A. E., & Johnson, B. (1997). Therapists' perspectives of couple problems and treatment issues in couple therapy. Journal of Family Psychology, 11 (3), 361-366.

APPENDIX A.

Permission is granted to utilize the Weiss Partner Betrayal Trauma Scale for counseling and research purposes with citation. For more information, please go to www.partnerbetrayaltrauma.org.

• 1. "My partner has engaged in sex acts of any kind with a close friend, neighbor, coworker, or acquaintance of mine."

Never Don't Know Once/Rarely Sometimes


• 2. "My partner has engaged in sex acts of any kind with someone I do not know."

Never Don't Know Once/Rarely Sometimes

Frequently

• 3. "My partner has paid for live sex services of any kind (e.g. prostitute, stripper)."

Never Don't Know Once/Rarely Sometimes

Frequently

• 4. "My partner makes commitments to me that he does not follow through with.".

Never Don't Know Once/Rarely Sometimes

Frequently

• 5. "My partner has viewed porn in secret"

Never Don't Know Once/Rarely Sometimes

Frequently

• 6. "My partner has masturbated in secret."
• 13. "My partner has avoided emotional intimacy with me."

Never Don't Know Once/Rarely Sometimes

Frequently

• 7. "My partner has participated in any type of online sexual relationship."

Never Don't Know Once/Rarely Sometimes

Frequently

• 8. "I believe my partner has a sex addiction of some type."

Never Don't Know Once/Rarely Sometimes

Frequently

• 9. "My partner has had inappropriate emotional relationships with others."

Never Don't Know Once/Rarely Sometimes

Frequently

• 10. "My partner has been immature emotionally."

Never Don't Know Once/Rarely Sometimes

Frequently

• 11. "My partner has challenges appropriately expressing anger"

Never Don't Know Once/Rarely Sometimes

Frequently

• 12. "My partner has been verbally abusive with me."

Never Don't Know Once/Rarely Sometimes

Frequently

Never Don't Know Once/Rarely Sometimes


• 14. "My partner has avoided sexual intimacy with me."

Never Don't Know Once/Rarely Sometimes

Frequently

• 15. "My partner has been disconnected during sex."

Never Don't Know Once/Rarely Sometimes

Frequently

• 16. "My partner blames me before he takes responsibility for an issue."

Never Don't Know Once/Rarely Sometimes

Frequently

• 17. "My partner has been unduly critical of me."

Never Don't Know Once/Rarely Sometimes

Frequently

• 18. "My partner has avoided sharing his feelings with me."

• 19. "My partner has avoided apologizing to me when warranted."

Never Don't Know Once/Rarely Sometimes

Frequently

Scoring Instructions: All Items are scored the same. Never = 0, Don't Know =1, Once/Rarely =2, Sometimes = 3, and Frequently =4. Scores can range from 0-76. Suggested cutoffs are:

46-60 = Very Substantial Betrayal Trauma Level

61 and above = Extremely Substantial Betrayal Trauma Level

Generated with Reader Mode