%20-%20https://sketch.com%20--%3e%3ctitle%3e3ECBBF50-FD08-412A-943D-2CFC5E025A16%3c/title%3e%3cdesc%3eCreated%20with%20sketchtool.%3c/desc%3e%3cg%20id='Stylesheet'%20stroke='none'%20stroke-width='1'%20fill='none'%20fill-rule='evenodd'%3e%3cg%20transform='translate(-926.000000,%20-668.000000)'%20id='BUTTONS'%3e%3cg%20transform='translate(100.000000,%20597.000000)'%3e%3cg%20id='BTN-/-RIGHTS'%20transform='translate(816.000000,%2065.000000)'%3e%3cg%20id='ICON-/-Rights'%20transform='translate(10.000000,%206.000000)'%3e%3cpath%20d='M13.5752841,0%20L1.34943182,0%20C0.59375,0%200,0.586666667%200,1.33333333%20L0,22.6666667%20C0,23.4133333%200.620738636,24%201.34943182,24%20L17.6505682,24%20C18.40625,24%2019,23.3866667%2019,22.6666667%20L19,5.36%20L13.5752841,0%20Z%20M13.5752841,1.89333333%20L17.0838068,5.36%20L13.5752841,5.36%20L13.5752841,1.89333333%20Z%20M1.34943182,22.6666667%20L1.34943182,1.33333333%20L12.1988636,1.33333333%20L12.1988636,5.36%20C12.1988636,6.10666667%2012.7926136,6.69333333%2013.5482955,6.69333333%20L17.6235795,6.69333333%20L17.6505682,22.6666667%20L1.34943182,22.6666667%20Z'%20id='Shape'%20fill='%23595959'%20fill-rule='nonzero'%3e%3c/path%3e%3ccircle%20id='Oval'%20stroke='%23595959'%20stroke-width='1.5'%20fill='%23595959'%20cx='9.5'%20cy='13.5355339'%20r='5.5'%3e%3c/circle%3e%3cpolyline%20id='Path-2'%20stroke='%23FFFFFF'%20stroke-width='1.5'%20transform='translate(9.500000,%2013.035534)%20rotate(45.000000)%20translate(-9.500000,%20-13.035534)%20'%20points='8%2016.0355339%2011%2016.0355339%2011%2010.0355339'%3e%3c/polyline%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
There is a growing body of scientific evidence indicating that individuals who have experienced psychological abuse over a period of time may suffer from mental health problems and other adjustment difficulties (e.g., Ansara & Hindin, Reference Ansara and Hindin2011; Aronoff, Lynn, & Malinoski, Reference Aronoff, Lynn and Malinoski2000). The negative consequences of psychological abuse can persist years after the abusive situation has remitted, and they can be even more severe than the effects of physical abuse (Street & Arias, Reference Street and Arias2001). Several studies have examined psychopathological symptoms through standardized measures in survivors of different abusive contexts, including intimate partner violence (Beck et al., Reference Beck, McNiff, Clapp, Olsen, Avery and Hagewood2011), bullying (Duarte, Pinto-Gouveia, & Rodrigues, Reference Duarte, Pinto-Gouveia and Rodrigues2015), elder abuse (Dong, Reference Dong2015), and abusive groups (Malinoski, Langone, & Lynn, Reference Malinoski, Langone and Lynn1999). However, restricting the effects of psychological abuse to psychopathological symptoms is unlikely to capture the full range of difficulties resulting from interpersonal abusive experiences (Rogers & Follingstad, Reference Rogers and Follingstad2014). Thus, psychological and social difficulties that usually do not reach a clinical significance to be considered mental health disorders also need to be taken into account, especially due to their severe impact on survivors’ daily lives (e.g., Ansara & Hindin, Reference Ansara and Hindin2011; Durocher, Reference Durocher1999).
Researchers have extensively documented different kinds of difficulties in individuals who have suffered psychologically abusive behaviors within social groups, organizations, or alternative communities with cultic dynamics (e.g., Coates, Reference Coates2012; Malinoski et al., Reference Malinoski, Langone and Lynn1999). These groups are usually labeled high-demand groups, manipulative groups, or abusive groups. Abusive groups have been defined as any group or movement of any kind that exhibits great or excessive devotion or dedication to a person, idea, or thing, and employs unethically manipulative persuasion and control practices designed to foster submission and advance the goals of the group’s leaders, to the current or possible detriment of members, their families, or the community (West & Langone, Reference West and Langone1986). The concept of group psychological abuse was proposed to describe the practices that characterize abusive groups, and it is defined as a process of systematic and continuous application of pressure, control, manipulation, and coercion strategies aimed to dominate and achieve the submission of the group’s members (Rodríguez-Carballeira et al., Reference Rodríguez-Carballeira, Saldaña, Almendros, Martín-Peña, Escartín and Porrúa-García2015). Examples of these strategies include isolation, intimidation, contempt, manipulation of blame, and control over one’s affective relationships. It has been argued that these abusive strategies result in some degree of psychological distress in members of abusive groups who would be relatively healthy without their group experiences (Almendros, Carrobles, Rodríguez-Carballeira, & Gámez-Guadix, Reference Almendros, Carrobles, Rodríguez-Carballeira and Gámez-Guadix2009; Singer & Ofshe, Reference Singer and Ofshe1990).
The available scientific literature on the phenomenon has extensively addressed the adjustment problems that survivors of abusive groups may experience while becoming integrated into the outside world. These problems include psychopathological symptoms such as anxiety, depression, dissociation, or posttraumatic stress disorder (Gasde & Block, Reference Gasde and Block1998; Malinoski et al., Reference Malinoski, Langone and Lynn1999; Martin, Langone, Dole, & Wiltrout, Reference Martin, Langone, Dole and Wiltrout1992). Furthermore, the long-term consequences of group psychological abuse include high levels of psychological stress (Saldaña, Rodríguez-Carballeira, Almendros, & Guilera, Reference Saldaña, Rodríguez-Carballeira, Almendros and Guilera2018) and a wide variety of psychological and social disturbances such as anger, low self-confidence, guilt, fear, mistrust, lack of social skills, and stigmatization (Boeri, Reference Boeri2002; Coates, Reference Coates2010; Matthews & Salazar, Reference Matthews and Salazar2014).
A comprehensive taxonomy of these psychological and social disturbances was recently developed and then validated using a panel of experts who found the emotional difficulties to be the most frequent and intense disturbances in survivors of abusive groups (Saldaña, Antelo, Rodríguez-Carballeira, & Almendros, Reference Saldaña, Antelo, Rodríguez-Carballeira and Almendros2018). The taxonomy included six type of disturbances that were operationally defined (Table 1).
Emotional Difficulties in Survivors of Abusive Groups
The existence of these emotional disturbances has been shown in qualitative, clinical, and survey-based studies. Specifically, Coates (2010) interviewed 9 individuals who had been involved in different abusive groups during adulthood and reported feelings of loss, shame, guilt, and low self-esteem. Similar results were found by Matthews and Salazar (Reference Matthews and Salazar2014), who interviewed 15 individuals who had been raised within an abusive group and also reported emotional difficulties such as fear and rage. Thus, according to qualitative studies, emotional difficulties are experienced by both first-generation (people who join the group during adulthood) and second-generation (people who were born or raised within the group) former members of abusive groups. Using well-validated instruments on distress and personality disorders, Martin et al. (Reference Martin, Langone, Dole and Wiltrout1992) found that more than 50% of their 308 participants reported experiencing anxiety, fear, worry, guilt, despair, and anger toward the group leader.
Although these studies found that former members of abusive groups reported significant clinical symptoms, only a few studies have examined the possible relationship between these symptoms and the practices endured by survivors of these groups (Saldaña, Rodríguez-Carballeira, Almendros, & Guilera, Reference Saldaña, Rodríguez-Carballeira, Almendros and Guilera2018). With these or other similar instruments, the results have been contradictory. Some studies have found a significant association between reported psychological abuse while in the group and current distress symptomatology (e.g., Winocur, Whitney, Sorensen, Vaughn, & Foy, Reference Winocur, Whitney, Sorensen, Vaughn and Foy1997), whereas others have not (e.g., Gasde & Block, Reference Gasde and Block1998). Thus, the next step in increasing our understanding of the long-term emotional consequences of group psychological abuse would be to develop specific instruments to measure the emotional disturbances commonly experienced by survivors of abusive groups.
The purpose of this study was to develop and examine the psychometric properties of a new measure of the emotional distress experienced by survivors of abusive groups. The four specific aims established were to analyze: (a) The internal structure and its replicability across sex and age of involvement in the group, (b) the internal consistency of the scores, (c) the discriminating power, and (d) the relation with group psychological abuse, psychopathological symptoms, trauma, and self-esteem.
Method
Participants
The study included 706 people from 18 to 78 years old (Age: M = 40.02, SD = 14.52; Women: 57 %, Men: 43 %). Their native language was Spanish, and they were mainly from Spain (68.7 %), whereas smaller percentages were from Latin America (27.8 %) or other European countries (2.7 %). Participants were asked to report their experiences with a group they had been members of in the past, but no longer belonged to when the study took place. If they had belonged to more than one group, participants were asked to select the group they now think was the most controlling towards its members. Participants belonged to groups that were mainly of a religious, personal development, commercial, or philosophical nature.
We distributed participants into two different samples according to whether or not they reported having experienced psychological abuse within the group they selected. A first sample was composed of 413 victims, and a second sample was composed of 293 non-victims. The sample of victims consisted of people who reported having suffered abusive behaviors (e.g., isolation, control over personal life, emotional abuse, or denigration of critical thinking) in a diverse and intense way that can be considered psychological abuse, and not merely socially-accepted influence, according to the optimal cut-off point on a scale measuring experiences of group psychological abuse –see instruments section–.
Taking into account the sample of victims, we also distinguished participants according to their age of involvement in the abusive group. Thus, 237 victims reported joining the group during adulthood, and they were considered first generation members, whereas 135 reported that they were born or raised within the group, and they were considered second-generation former members. Table 2 shows the main descriptive sociodemographic and group-related information for the sample of victims and non-victims, as well as the reported degree of group psychological abuse experienced and the different types of support received by participants in relation to their group experiences.
Descriptive Data of the Samples of Victims and Non-Victims
Note: Victims n = 413; Non-victims n = 293. Group psychological abuse = Measured through the Psychological Abuse Experienced in Groups Scale.
Instruments
The Emotional Distress Scale in Survivors of Abusive Groups (EDS-SAG) is a self-reported questionnaire aimed to assess the degree of emotional distress experienced by survivors of group psychological abuse. It was developed following the guidelines revised by Kline (Reference Kline2015), and using the semantic definitions of emotional disturbances proposed by Saldaña et al. (2018). First, four researchers with expertise in the negative impact of group psychological abuse elaborated an initial pool of 73 items. These items were reviewed by an external panel of nine experts from Spain to assure its content validity. Experts were asked to evaluate the relevance, clarity, and representativeness of each item. Based on their assessments, 35 items were selected and administered to a pilot sample composed of 14 survivors of abusive groups. Their responses were analyzed in order to retain the items with better properties from both a qualitative and quantitative perspective. The final version of the EDS-SAG (see Appendix) included 18 items, and the response format for the items is a 5-point Likert-type scale (0 = not at all; 1 = slightly; 2 = moderately; 3 = strongly; 4 = very strongly). The response labels were selected to assess the intensity with which each emotional disturbance was experienced by the respondent.
The Psychological Abuse Experienced in Groups Scale (PAEGS) (Saldaña, Rodríguez-Carballeira, Almendros, & Escartín, Reference Saldaña, Rodríguez-Carballeira, Almendros and Escartín2017; Spanish adaptation by Saldaña, Rodríguez-Carballeira, & Almendros, Reference Saldaña, Rodríguez-Carballeira and Almendros2018) was used to assess the degree of perceived group psychological abuse experienced while in the group, in order to provide evidence of the relation with a relevant external variable and to distribute participants into the samples of victims and non-victims. The PAEGS is a self-report questionnaire composed of 31 items rated on a 5-point Likert-type scale (0 = not at all; 1 = slightly; 2 = quite a lot; 3 = a lot; 4 = continually). Previous studies with Spanish-speaking survivors of abusive groups reported consistent evidence of the one-dimensional structure, adequate reliability of the scale scores and discriminatory power, and its relationship with external variables. A score above 27 has been found to be useful as a threshold for detecting group psychological abuse experiences in the Spanish-speaking population, showing a sensitivity of 94.4% and a specificity of 99.3% (Saldaña, Rodríguez-Carballeira, & Almendros, Reference Saldaña, Rodríguez-Carballeira and Almendros2018). This empirical criterion was used to classify participants in the current study into the two samples. As in prior studies, we found adequate internal consistency coefficients for the overall score, both in the sample of victims (ω = .97) and in the sample of non-victims (ω = .81).
The Brief Symptom Inventory (BSI) (Spanish adaptation by Ruipérez, Ibáñez, Lorente, Moro, & Ortet, Reference Ruipérez, Ibáñez, Lorente, Moro and Ortet2001) was used to assess possible current psychopathological symptoms in order to provide evidence of relation with clinical variables. The BSI is a widely used instrument composed of 53 items rated on a 5-point Likert-type scale ranging from 0 (never) to 4 (very often). Respondents were asked to rate the extent to which each identified problem had caused them discomfort in the past week. In this study, we considered the Global Severity Index and the nine symptom dimensions provided by the BSI, i.e. Anxiety, Depression, Hostility, Interpersonal Sensitivity, Obsessive-Compulsive, Paranoid Ideation, Phobic Anxiety, Psychoticism, and Somatization. McDonald’s Omega coefficients for these symptom dimensions ranged from .86 (Paranoia Ideation) to .93 (Depression).
The Posttraumatic Cognitions Inventory (PTCI) (Spanish adaptation by Blanco, Díaz, Gaborit, & Amaris, Reference Blanco, Díaz, Gaborit and Amaris2010) was used to evaluate trauma-related thoughts and beliefs and provide further evidence of relation to clinical variables. It is composed of 36 items rated on a 7-point Likert-type scale ranging from 1 (totally disagree) to 7 (totally agree). The PTCI provides a total score and scores for three dimensions measuring Negative cognitions about the self, Negative cognitions about the world, and Self-blame. In the present study, their McDonald’s Omega coefficients ranged from .75 (Self-blame) to .95 (Negative cognitions about the self).
The Rosenberg Self-Esteem Scale (RSES) (Spanish adaptation by Martín-Albo, Núñez, Navarro, & Grijalvo, Reference Martín-Albo, Núñez, Navarro and Grijalvo2007) was administered to examine current personal self-esteem, understood as feelings of self-respect and self-acceptance. The RSES includes 10 items rated on a 4-point Likert-type scale ranging from 1 (totally disagree) to 4 (totally agree). In the present study, the McDonald’s Omega coefficient was .91.
Procedure
The current study was approved by the University of Barcelona Bioethics Commission. Data collection took place between August and November 2015 through an online questionnaire using convenience non-probabilistic and snowball sampling methods. In order to contact potential victims of group psychological abuse, the study was announced mainly through victim support associations and other organizations that provide information, education, and counseling about abusive groups, as well as through mental health professionals who work with this population, specialized online forums of former members of abusive groups, and other previous participants in the study. In order to contact potential former members of non-abusive groups, we announced the study through mainstream society organizations and social networks. A description of the study and the link to the questionnaire were published on the websites, forums, and other relevant social networks of the organizations and professionals who collaborated in the study. All the participants were informed about the goals of the study, gave their informed consent, and collaborated voluntarily and without receiving compensation.
Data Analysis
After checking that there were no missing data, an exploratory factor analysis was conducted, taking into account the victims’ scores to examine the internal structure of the EDS-SAG using FACTOR 9.3. The unweighted least squares (ULS) extraction method was used with the polychoric correlation matrix, due to its robustness with small samples and Likert-type items. The information provided by the parallel analysis and the results of the Hull method were taken into account to select the number of factors. The goodness-of-fit of the data to the model was established through the goodness-of-fit index (GFI) and the root mean square of residuals (RMSR).
Replication analyses were conducted following the guidelines proposed by Osborne and Fitzpatrick (Reference Osborne and Fitzpatrick2012) in order to explore the stability of the factorial solution across relevant biographical variables. In this regard, we conducted further EFA analyses, splitting the sample of victims according to sex and the participants’ age of involvement in the abusive group (i.e. first-generation and second-generation former members). Then, the resulting factor loadings and structures were compared to test the replicability of the EDS-SAG.
The internal consistency of the scores on the measures included in the study was examined by computing McDonald’s Omega coefficient (ω), which seems to be one of the best alternatives for estimating reliability (Revelle & Zinbarg, Reference Revelle and Zinbarg2009). Cliff’s delta (d) coefficient was used to examine the effect size of the differences between samples on the EDS-SAG score, due to its robustness with non-normal and ordinal data. The discriminating power of the EDS-SAG was examined by means of the receiver operating characteristic (ROC) curve procedure using MedCalc.14. Evidence of the relationship with other relevant variables was explored with bivariate correlation analyses between the EDS-SAG scores and scores on the PAEGS, BSI, PTCI, and RSES. Finally, the differences in reported emotional distress, based on the main sociodemographic variables, were analyzed using the Mann-Whitney U test, due to the non-normal distribution of the scores.
Results
Factor Structure Analysis
The factor structure of the EDS-SAG scores was tested using the data from the sample of victims. Results of the Kaiser-Meyer-Olkin index (.93) and Bartlett’s sphericity test (χ2 = 3650.4, p < .001) confirmed the adequacy of the current matrix for factor analysis. Parallel analysis and the Hull method recommended the extraction of one major factor with a total explained variance of 50.7%. In addition, the GFI (.98) and the RMSR (.07) also supported the one-factor structure, showing a good fit of the data. As Table 2 shows, factor loadings ranged from .62 to .79, in all cases exceeding the .4 criterion that leads to including an item in the interpretation of a factor.
Table 3 shows the descriptive properties of the 18 items on the EDS-SAG, calculated from the responses of the sample of victims. Most of the items presented negative skewness, and the corrected item-total correlation was higher than the .4 criterion in all cases. The McDonald’s Omega coefficient was .94, showing an appropriate internal consistency of the EDS-SAG scores in the sample of victims.
Descriptive Statistics of the EDS-SAG Item Scores for the Sample of Victims
Note: n = 413; 95% CI = 95% confidence interval; r cix = corrected item-total correlation score; λi1 = item’s factor loadings.
Replicability Analysis
The replicability of the EDS-SAG was tested by conducting four exploratory factor analyses, splitting the sample of victims first by sex (i.e. women and men) and then by the age of involvement in the abusive group (i.e. first-generation and second-generation former members). Results of the Kaiser-Meyer-Olkin indexes (Women: .91, Men: .92, First-generation: .92, Second-generation: .87) and the values of Bartlett’s Sphericity tests (Women: χ2 = 1,815, p < .001, Men: χ2 = 1,954, p < .001, First-generation: χ2 = 2,079.5, p < .001, Second-generation: χ2 = 1,155.2, p < .001) confirmed the adequacy of the matrices for factor analysis. Parallel analysis and the Hull method recommended the extraction of one major factor in all cases. The extracted factor explained a total variance of 50% in the subsamples of women and men, 49% in the subsample of first-generation former members, and 45% in the subsample of second-generation former members. The GFI (Women: .98, Men: .98, First-generation: .98, Second-generation: .96) and the RMSR (Women: .07, Men: .08, First-generation: .08, Second-generation: .09) also supported one-factor structures, showing a good fit of the data. Finally, regarding the squared differences in the factor loadings, none of them achieved a magnitude of .04, which is a common criterion to view a factor loading as volatile (Osborne & Fitzpatrick, Reference Osborne and Fitzpatrick2012).
Replicability Analyses of the EDS-SAG Internal Structure
Note: n = 413; λi1 = item’s factor loading; Difference = Squared differences.
Discriminatory Power
The theoretical range of scores on the EDS-SAG is between 0 and 72. The sample of victims obtained an average score of 38.97, 95% CI [37.30, 40.64]; SD = 17.29, and the sample of non-victims obtained an average score of 1.68, 95% CI [1.14, 2.22]; SD = 4.69. Significant rank differences with a high magnitude were obtained between the two samples (U = 2,037.00, p < .001; Cliff’s d = .96). Results of the ROC curve analysis also supported a high discriminative capacity of the EDS-SAG to distinguish between the degree of emotional distress experienced by survivors of abusive groups and the degree experienced by former members of non-abusive groups, with the area under the curve obtaining a value of .983, 95% CI [.971, .991], p < .001.
Relation with External Variables
Bivariate correlations between the EDS-SAG scores and construct-related measure scores were examined to provide evidence of their relationships. As in previous studies, the responses of the samples of victims and non-victims were taken into account together to increase the variability of the measures (Saldaña, Rodríguez-Carballeira, & Almendros, Reference Saldaña, Rodríguez-Carballeira and Almendros2018). On the one hand, a significant high-magnitude positive correlation was found between the EDS-SAG and the PAEGS (.86, p < .001). On the other hand, significant correlations were also found between the EDS-SAG scores and the psychological distress measures. First, both the Global Severity Index (.30, p < .001) and the nine symptom dimensions of the BSI correlated with the EDS-SAG scores: Paranoia Ideation (.32, p < .001), Depression (.28, p < .001), Interpersonal Sensitivity (.28, p < .001), Anxiety (.26, p < .001), Obsessive-Compulsive (.24, p < .001), Psychoticism (.23, p < .001), Phobic Anxiety (.23, p < .001), Somatization (.19, p < .001), and Hostility (.16, p < .001). Second, a significant negative correlation was also found between the EDS-SAG scores and the RSES scores (–.23, p < .001). Third, significant correlations that were medium in magnitude were also found with the PTCI total score (.46, p < .001), as well as with its three subscales: Negative cognitions about the self (.42, p < .001), Negative cognitions about the world (.37, p < .001), and Self-blame (.37, p < .001).
Relation with Biographical Variables
Regarding the scores obtained on the EDS-SAG in the sample of victims, significant differences were found based on sex (Women: M = 41.35, SD = 17.07; Men: M = 36.63; SD = 17.26; U = 1,7843, p = .005; d = .16), and on the age of involvement in the abusive group (First-generation: M = 39.16, SD = 16.73; Second-generation: M = 43.16; SD = 16.01; U = 1,3822, p = .029; d = –.14). However, no significant correlation was found between the EDS-SAG scores and time spent inside the group (.01, p = .76), and a negative significant correlation, although low in magnitude, was found with time outside the group (–.15, p < .001). To interpret the scores on the EDS-SAG, Table 5 provides the raw scores and their corresponding percentiles in the sample of victims. For the differences found in terms of sex and generation, separate scores are provided to allow a simple and adequate interpretation.
Percentiles of the EDS-SAG Scores for the Sample of Victims
Note: First generation: Women n = 121, Men n = 116; Second generation: Women n = 61, Men n = 73.
Discussion
The purpose of this study was to examine the psychometric properties of the EDS-SAG scores in a Spanish-speaking sample. This new instrument was intended to assess the degree of emotional difficulties that survivors of abusive groups may suffer, including the most common and intensely experienced disturbances among survivors, according to experts in the area (Saldaña et al., 2018). The items on the EDS-SAG cover emotional difficulties that are widely reported in the scientific literature, such as fear, shame, feelings of loss, despair, and rage (Boeri, Reference Boeri2002; Coates, Reference Coates2010; Matthews & Salazar, Reference Matthews and Salazar2014). The use of a comprehensive semantic definition of emotional difficulties as a starting point, and the initial evaluation of the items by an external panel of experts, were key aspects in ensuring content validity.
Regarding the internal structure of the scores of the EDS-SAG, results of the factor analysis supported a unidimensional solution, with one common factor explaining more than 50% of the total variance. This finding suggests that the emotional disturbances represented in the items on the scale and included in the taxonomy provided by Saldaña et al. (2018), are often experienced simultaneously and to a similar degree by survivors of group psychological abuse. Moreover, replicability analyses showed that the unidimensional structure of the scale remains stable across sex (i.e. women and men) and age of involvement (i.e. first-generation and second-generation), thus providing more robustness to the interpretation of the total scale score when assessing emotional disturbances in survivors of abusive groups. In this regard, our findings are consistent with evidence found in qualitative and survey-based studies, some using ad-hoc items to address emotional disturbances, where survivors of abusive groups, both women and men, as well as first-generation and second-generation former members, reported having experienced diverse emotional disturbances (e.g., Coates, Reference Coates2010, Matthews & Salazar, Reference Matthews and Salazar2014). Furthermore, reliability analysis showed adequate internal consistency of the scores on the EDS-SAG obtained by victims of group psychological abuse because McDonald’s Omega coefficient exceeded the recommended criterion when working with psychological measurements (Reise, Bonifay, & Haviland, Reference Reise, Bonifay and Haviland2013).
Results of the discriminatory analysis showed that the EDS-SAG has the capacity to distinguish the emotional disturbances suffered by survivors of abusive groups from those of people who have never been involved in such groups. Thus, the results of the ROC curve analysis, along with the high magnitude differences between victims and non-victims in the scores on the EDS-SAG, suggest that the proposed scale evaluates emotional disturbances specifically experienced by survivors of abusive groups.
Regarding evidence of the relationship with construct-related variables, a high magnitude correlation was found between the EDS-SAG and the PAEGS scores. In this regard, the degree of emotional distress is closely related to the degree of psychological abuse participants reported having experienced while in the group. These results indicate that some abusive strategies, such as isolation from the social support network, humiliation, intimidation, or manipulation of guilt, may foster post-involvement emotional disturbances such as grief, guilt, fear, or low self-esteem, as several authors have suggested (e.g., Rodríguez-Carballeira et al., Reference Rodríguez-Carballeira, Saldaña, Almendros, Martín-Peña, Escartín and Porrúa-García2015). An interesting finding is that the extent of emotional distress was not correlated with time spent within the group. Thus, even when abusive practices in these groups are experienced for a relatively short time, their impact on an emotional level can be equally severe in survivors of abusive groups. It is also worth noting that the correlation coefficient found in the current study between the PAEGS and the EDS-SAG was higher than those found in most previous studies examining relations between group psychological abuse and other distress measures (e.g., Winocur et al., Reference Winocur, Whitney, Sorensen, Vaughn and Foy1997; Wolfson, Reference Wolfson2002). Therefore, the EDS-SAG seems to evaluate the specific distress suffered by this type of victims better than other instruments aimed toward the general population that do not capture the specific negative impact of victimization experiences.
Regarding other evidence of the relationship with other clinical variables, a higher prevalence of emotional distress measured by the EDS-SAG was significantly correlated with all the mental health outcome indicators. The moderate correlations found in the current study suggest that the emotional distress measured by the EDS-SAG and the distress measured by the BSI, the PTCI, and the RSES are related constructs. However, whereas the BSI and other distress instruments evaluate disturbances in a general way, without taking into account the characteristics of the studied population, the EDS-SAG assesses the specific distress experienced by a particular population of victims. The items on the EDS-SAG are written in a way that better reflects the experiences of survivors of group psychological abuse. In any case, based on the correlations between the scores on the distress measures in the current study, the results suggest that survivors of group psychological abuse may present a complex clinical picture characterized by emotional difficulties, an overall negative self-evaluation, feelings of personal inadequacy and inferiority, symptoms of depression and anxiety, and feelings of self-blame (e.g., Coates, Reference Coates2010; Martin et al., Reference Martin, Langone, Dole and Wiltrout1992). Our findings suggest that the degree of emotional distress may be reduced over time once outside the group, although the current data do not allow us to specify which personal or social factors contribute to this reduction.
Finally, it is important to highlight that the emotional distress evaluated must be interpreted in terms of sex and generation. Considering the results obtained through the replication analysis, the differences observed are not due to a different functioning of the scale structure. Results showed that women suffer greater emotional distress than men, which is consistent with the findings of qualitative (Boeri, Reference Boeri2002) and quantitative studies using other instruments to measure distress (e.g., Almendros et al., Reference Almendros, Carrobles, Rodríguez-Carballeira and Gámez-Guadix2009). In terms of whether the survivors were born and/or raised within the group or not, results showed that second-generation former members suffer greater emotional distress than first-generation former members. One explanation for this finding is that psychological abuse in early childhood has been linked to attachment disorders, developmental and educational problems, socialization problems, and disruptive behavior (Hibbard et al., Reference Hibbard, Barlow and MacMillan2012), which can also exacerbate emotional disturbances. Second-generation and first-generation former members may experience a similar degree of psychological abuse (Saldaña et al., Reference Saldaña, Rodríguez-Carballeira, Almendros and Escartín2017), although the impact of such abuse on the emotional wellbeing of people raised within the group can be more pronounced (Matthews & Salazar, Reference Matthews and Salazar2014).
This study has relevant strengths, but also some limitations. First, the representativeness of the sample could not be verified, given the difficulty of accessing survivors of abusive groups, which may be considered a hard-to-reach and hidden population (Shaghaghi, Bhopal, & Sheikh, Reference Shaghaghi, Bhopal and Sheikh2011). However, the sample size of survivors of group psychological abuse in this study is superior to previous studies, which were usually composed of about 100 participants. Moreover, it is not a clinical sample consisting only of people in treatment with researchers or related professionals. A second limitation is related to the self-report nature of the EDS-SAG and the retrospective evaluation of abusive experiences, another common issue in studies designed to assess interpersonal violence (Almendros, Carrobles, Rodríguez-Carballeira, & Jansà, Reference Almendros, Carrobles, Rodríguez Carballeira and Jansa2004). Another consideration would be the differences in demographic variables between the samples. However, the non-victim sample in the current study is not limited to university students (e.g., Saldaña, Rodríguez-Carballeira, Almendros, & Nishida, Reference Saldaña, Rodríguez-Carballeira, Almendros and Nishida2018) or to only former members of religious groups (e.g., Gasde & Block, Reference Gasde and Block1998).
Despite the aforementioned considerations, the development of the EDS-SAG represents a significant step forward in the study of the long-term negative consequences of psychological abuse. The results obtained in this study provide enough empirical evidence to support the use of the EDS-SAG for research purposes, facilitating the assessment of emotional distress experiences in survivors of abusive groups. In this regard, the EDS-SAG provides new opportunities to evaluate the antecedents of emotional distress, its protective factors, and its relationship with clinical symptomatology using standardized measures. Moreover, in clinical contexts, the EDS-SAG and its corresponding percentiles could be useful for professionals as a screening tool to provide preliminary information at initial assessment or to examine the efficacy of therapeutic interventions with survivors of group psychological abuse.
The current study also points to new directions for future research in this area. First, it would be necessary to continue to explore the psychometric properties of the EDS-SAG using new samples from different cultural contexts. In addition, it would be interesting to more deeply assess the role of emotional disturbances in the development and maintenance of psychopathological symptoms and stress responses, as well as examining which factors contribute to reducing emotional distress once survivors are outside the group, such as social support, copying strategies, or psychological counselling. Moreover, future research should continue to develop new measurement instruments to evaluate other psychological and social difficulties in survivors of abusive groups, such as those related to the cognitive and relational domains, as well as in survivors of other contexts of interpersonal violence. All of this would help to better understand the long-term consequences of psychological abuse, which is essential for their identification, treatment, and prevention in clinical settings.
Appendix
Emotional Distress Scale in Survivors of Abusive Groups (EDS-SAG)
