Psychometric Properties of the Checklist of Experiences of Psychological Abuse to the Couple in Adolescents and Young Adults

Abstract

In Ibero-America, interest in the evaluation and prevention of violence in young couples has been increasing. However, there are not enough Spanish-speaking standardized questionnaires. The psychometric analyses of the Checklist of Experiences of Psychological Abuse to the Couple (CEPA) are presented as a subset of the Checklist of Experiences of Abuse to the Couple. Overall, 1,505 adolescents participated; 828 were women, and 677 were men, aged 14 to 21 years old. A confirmatory factor analysis with the Structural Equation Modeling Software version 6.2 (EQS) was performed to consider different factor models that were consistent with theoretical and empirical aspects about violence in young couples. A structure was obtained with one factor of higher order, “Psychological Aggression” (14 items), and the following three primary factors: (a) “Coercion” (four items), (b) “Humiliation” (five items), and (c) “Control” (five items); Cronbach’s alpha ranged from .71 to .83. The overall scale and the three subscales correlated significantly with each other and with negative personality traits associated with gender. The results indicate that the scale has adequate psychometric properties for use in clinical practice and research.

Keywords

dating violence confirmatory factor analysis reliability adolescents young adults

Violence in young unmarried couples has received much attention in recent years due to the high prevalence rates reported by several studies in different countries (Rubio-Garay, López-González, Carrasco, & Amor, 2017). The prevalence studies show that more than half of the adolescents and young people surveyed were victims of some type of abuse exercised by their partner (Bringas-Molleda et al., 2015; Farhat, Haynie, Summersett-Ringgold, Brooks-Russell, & Iannotti, 2015; Garrido & Taussig, 2013; Shorey, Seavey, Brasfield, Febres, Fite, & Stuart, 2015; Williams, Miller, Cutbush, Gibbs, Clinton-Sherrod, & Jones, 2015), with the most frequent type being psychological abuse (PA), followed by physical and sexual violence (López-Cepero Borrego, Rodríguez-Franco, Rodríguez-Díaz, Bringas-Molleda, & Paino, 2015; Reidy, Smith-Darden, Cortina, Kernsmith, & Kernsmith, 2015). A systematic review conducted with 113 prevalence studies selected among 1,221 initial studies, retrieved from databases such as PsycINFO, Medline, and Scopus, found that PA was the most common, followed by physical and sexual violence. The rates reported by these studies ranged between 4.2% and 97% for the PA committed, and between 8.5% and 95.5% for the PA suffered (Rubio-Garay et al., 2017).

In Ibero-America, there have been some prevalence studies that reveal high rates of violence in young unmarried couples, whose data show the need to create alternatives for the assessment, treatment, and prevention in the countries of the region (e.g., Almendros, Gámez-Guadix, Carrobles, Rodríguez-Carballeira, & Porrúa, 2009; Cortés-Ayala et al., 2014; Lila, Gracia, & Murgui, 2013; Vizcarra Larrañaga & Poo, 2011). In Colombia, for example, there have been two published studies about the prevalence of abuse in dating. The first one found that 82.6% of 403 university students between 15 and 30 years old had at least once experienced some form of abuse from their partner (Rey-Anacona, 2009), the PA is considered the most prevalent aspect (82.6% men and 80.3% women), followed by physical abuse (26.5% and 20%, respectively) and sexual abuse (10.8% and 6.8%, respectively). The second study examined the same population and revealed that 85.6% of 902 participants between 15 and 35 years old had been abusive at least once to their partner, with the most frequent type of abuse being psychological, followed by physical, emotional, sexual, economic, and negligent types of abuse (Rey-Anacona, 2013).

Despite the difficulties in physical and mental health associated with dating abuse (Centers for Disease Control and Prevention, 2006; Cortes-Ayala et al., 2014; Delgado-Alavarez, Estrada-Aranda, & López-Huerto, 2015; Herrero, Rodríguez-Díaz, & Torres, 2017), there has been relatively little progress regarding PA (Almendros et al., 2009). The relevance of PA has not been recognized until recently. However, it must be considered a predictor of physical aggression, and it has a comparable impact on mental health (self-esteem, anxiety, depression, etc.) and a more persistent impact over time (Herrero, Torres, & Rodríguez, 2018; Herrero, Torres, Rodríguez, & Juarros-Basterretxea, 2017). This long-term impact is because it exerts a key influence on the victim’s perception of their relationship, which then influences the decision-making process to leave the abusive situation (Almendros et al., 2009; Ha, Overbeek, Cillessen, & Engels, 2012; Messinger, Rickert, Fry, Lessel, & Davidson, 2012; Rodríguez-Franco et al., 2012; Straus & Douglas, 2004).

This consideration has so far led researchers to make various attempts to operationalize the PA construct, usually through the identification of typology (threats, insults, symbolic violence, humiliation, jealousy, control, etc.), without reaching a consensus about this (Rodríguez-Franco, López-Cepero Borrego, & Rodríguez-Díaz, 2009). The available definitions intermingle aspects concerning the consequences for the victim, the characteristics of the abusive act, and the intent of the perpetrator, leading Rodríguez-Carballeira et al. (2005) to propose a PA classification of the following six categories grouped into four dimensions: (a) contextual or situational: isolation, control, and manipulation of information and control of one’s own thinking; (b) emotional: emotional abuse; (c) cognitive: imposition of thought itself; and (d) behavioral: imposition of a subservient role (Almendros et al., 2009).

In actuality, the literature shows a clear difficulty to reach agreement on the components to assess PA, facilitating significant discrepancies and some ambiguity in research (Exner-Cortens, Gill, & Eckenrode, 2016a; López-Cepero Borrego, Rodríguez-Franco, & Rodríguez-Díaz, 2015). As a result, attempts to assess this construct remain scarce and inconsistent (Almendros et al., 2009). At the same time, the results of research show that it is necessary to have instruments collecting data on abuse in couples of teenagers and young adults, to identify cases in educational and health institutions, to evaluate interventions for victims and perpetrators of this form of violence, and to weigh the effectiveness of campaigns in primary and secondary prevention of this problem, which is considered a public health problem (Evers et al., 2013; Farhat et al., 2015; López-Cepero Borrego, Rodríguez-Franco, & Rodríguez-Díaz, 2015; Taylor, Calkins, Xia, & Dalla, 2017).

In Ibero-America, the phenomenon is increasingly being investigated. Already, some instruments have been adapted and/or validated to collect information on violence in young unmarried couples. Among international scales, the most used and most recognized is the Conflict Tactics Scale (CTS; Straus, 1979; Straus, Hamby, Boney-McCoy, & Sugarman, 1996), which evaluates different types of abuse, including verbal, minor physical, and severe physical aggressions. The modified version of this scale (M-CTS) was adjusted to Spanish by Muñoz-Rivas, Andreu-Rodríguez, Graña-Gómez, O’Leary, and González (2007), who have reported evidence in favor of the scale’s reliability and construct validity. However, this instrument does not assess the different forms of PA that could be present in youth dating relationships, as jealousy and control (Rodriguez-Franco et al., 2010).

Fernández-Fuertes, Fuertes, and Pulido (2006) adapted the Conflict in Adolescent Dating Relationships Inventory (CADRI; Wolfe et al., 2001), which allows individuals responding to the instrument to self-report whether they and/or their partner had exhibited any of a series of behaviors of the following five types of violence: (a) physical abuse, (b) verbal/emotional abuse, (c) sexual abuse, (d) threatening behavior, and (e) relational aggression (i.e., the one that seeks to affect interpersonal relationships or the reputation of the individual). They applied it to 572 Spanish men and women aged 15 to 19 years and obtained internal consistency indexes ranging from .51 for the threat scale and .79 for the verbal-emotional violence scale, with an alpha value of .86 for the entire questionnaire. The CADRI version translated into Spanish by Hokoda et al. (2006) presented alpha values ranging from .41 to .87 and test–retest indexes between .51 and .78. Although the CADRI measures different forms of PA that seem to be very frequent in dating violence, such as blaming, jealousy, ridiculing the other, threats, and attacking the couple’s social relationships, it does not present many items regarding control behaviors and emotional punishment, which could be very common in this type of violence (Rodriguez-Franco et al., 2010).

The “Questionnaire of Dating Violence” (Cuestionario de Maltrato en el Noviazgo [CMN]; Osorio-Guzmán, Tani, Bazan, Bonechi, & Menna, 2012) allows the report of different forms of physical, psychological, sexual, and economic partner abuse, using 52 items that are answered by a Likert-type scale of five options, ranging from “never” to “always.” This questionnaire also has a scale of 10 items assessing the “influence of socio-cultural values,” which features five response options ranging from “strongly agree” to “strongly disagree” and seeks to determine the presence of preconceived or discriminatory family roles toward the female gender and traditional attitudes toward gender roles, including variables such as the experience of violence in the family context and violent family patterns.

The “Questionnaire of Violence in Dating” (Cuestionario de Violencia entre Novios [CUVINO]; Rodríguez-Franco et al., 2010) was validated with a sample of 5,170 preuniversity and university students between the ages of 15 and 26 years, from Spain, Mexico, and Argentina. It has 41 items, with a Likert-type scale with five options of response, about behaviors belonging to the following eight types of maltreatment obtained through the factorial analysis: Detachment, Humiliation, Sexual, Coercion, Physical, Gender, Emotional and Instrumental Punishment. The eight factors correlated significantly with each other and explained 51.3% of the variance, with alpha indexes ranging from .58 to .81. This questionnaire better discriminates different types of abuse, including diverse forms of PA (detachment, humiliation, coercion, emotional and instrumental punishment; see also Rodríguez-Díaz et al., 2017, for analysis of a short-modified version of this instrument).

The Checklist of Experiences of Abuse to the Couple (Rey-Anacona, 2009) allows the report of the frequency of 95 different behaviors of physical, verbal, emotional, psychological, sexual, economic, and negligent abuse exercised toward the couple, through a Likert-type scale of four choices (from “never” to “often”); Rey-Anacona, Mateus, and Bayona (2010) reported a fairly high alpha (.96) in a sample of 562 men and women between 15 and 20 years, finding that 50.54% of the variance could be explained by six factors. Although the instrument had already been published in previous studies (Rey-Anacona, 2013, 2015), it was found to be very long and to require more evidence to its construct validity and criterion.

The instruments mentioned in this article showed as limitation the unidimensional evaluation of the PA construct, considering types of behavior such as insults, humiliation, jealousy, control, and others. Then, the Checklist of Experiences of Psychological Abuse to the Couple (CEPA) is a proposed short instrument to explain the PA construct as multidimensional bearing in mind three factors: control, humiliation, and coaction. The above mentioned allows to obtain a precise, reliable, and consistent measure because the PA construct is a prognosticator of the physical aggression and other kind of violence in dating.

This article presents the CEPA (Lista de Chequeo de Experiencias de Maltrato Psicológico en la Pareja), a short version of the Checklist of Experiences of Abuse to the Couple (Rey-Anacona, 2009), which seeks to overcome the limitations of the original instrument identifying its precision and reliability and the following limitations of previous research about instruments measuring PA in young unmarried couples in Ibero-America in an unidimensional way: (a) In general, these studies did not evaluate other forms of validity including the construct validity through factorial analysis, and this limitation was not exclusive to these studies, as Exner-Cortens, Gill, and Eckenrode (2016a; Exner-Cortens et al., 2016b) reported this limitation in their analysis of the psychometric properties of 13 instruments measuring behaviors and attitudes of dating violence; (b) the cultural context must be considered, assuming that it influences the different manifestations of violence, including dating violence; it is necessary to foresee differences in the manifestations of this violence as they apply in the countries where these instruments have been adapted or validated, but not in cultures such as Colombian; and (c) none of these instruments measured exclusively PA, the type of partner violence perhaps most difficult to measure due to the different positions that exist on its nature; therefore, more research is required that establishes and classifies behaviors that would be characteristic of this form of violence (Almendros et al., 2009). In the specific case of dating violence, this form of violence is of greater importance, bearing in mind that different prevalence studies indicate that it is more common than physical and sexual violence (e.g., Sears, Byers, & Price, 2007; Vizcarra Larrañaga & Póo, 2011). Research on the psychometric properties of these instruments is essential because evidence in favor of their validity and reliability is required for the study of the prevalence, correlates, possible consequences, and effectiveness of dating violence prevention alternatives (Exner-Cortens et al., 2016a).

Based on the above, the general objective was to evaluate the psychometric properties of the CEPA in a sample of Colombian adolescents and young adults, men and women, and the specific objectives were as follows: (a) to examine the factor and correlational structure of the instrument; (b) to evaluate its construct validity, subjecting the resulting correlation factors with scales from the Masculinity and Femininity Inventory (Inventario de Masculinidad y Femineidad [IMAFE]; Lara, 1993); (c) to analyze the internal consistency of the instrument and the resulting factors; and (d) to obtain scales overall and by sex, examining the differences between men and women overall and in each of the above factors.

Method

Participants

The analysis included the data of 1,505 teens and young people, 828 (55%) women and 677 (45%) men with ages ranging between 14 to 21 years old (M = 17.40 years, SD = 2.01). Some (885) of them were students of the Pedagogical and Technological University of Colombia (UPTC), 375 male (42.4%) and 510 female (57.6%), aged between 15 and 21 years (M = 8.66 years; SD = 1.38), who were enrolled in the first to 11 semester of any of the 23 academic programs offered by the University, with an average of 2.45 attended semesters (SD = 1.56). The remaining 620 were high school students enrolled in ninth, 10th, and 11th grade of three public schools in the city of Tunja (Boyacá), 318 (51.3%) female and 302 (48.7%) male, aged between 14 and 20 years old (M = 15.6 years; SD = 1.25). Participants were recruited through direct contact, according to their availability of participation. The inclusion criteria were as follows: had or having a romantic relationship, being single, and without children.

According to the classification of the National Administrative Department of Statistics (DANE), these people lived in neighborhoods of the following socioeconomic levels: extremely low (n = 305, 20%), low (n = 595, 39%), medium-low (n = 447, 29.7%), medium (n = 114, 7.6%), medium-high (n = 22, 1.5%), and high (n = 22, 1.5%; no data: 1.4%).

Instruments

Checklisted of experiences abuse to the couple

The items of this instrument are based on international and official definitions of the different types of Abuse to the Couple (Browne & Herbert, 1997; Colombian Institute of Legal y Ciencias Forenses.Medicine and Forensic, 2005; deMinistry of Health of Colombia, 1999), including six scales (see Rey-Anacona, 2009, 2013; Rey-Anacona et al., 2010). These items were inserted into the instrument, starting with the kind of violence containing the largest number of items and terminating with the fewest items in the following order: psychological, emotional, physical, sexual, economic, and neglected.

Once structured, the items were reviewed by three academic peers holding master’s degrees in social sciences with relevant research experience in Abuse to the Couple, according to the Administrative Department of Science, Technology and Innovation (Colciencias), and their observations were used to adjust the instrument until full agreement was reached.

The survey was submitted to a pilot test with a sample of college students aged between 18 and 23 years old, and other minor adjustments were made based on their observations, showing a satisfactory Cronbach’s alpha (>.90).

As previously announced, the CEPA constitutes a short version of the PA scale of this instrument, which understands PA as any action aimed at (a) controlling, restricting movements, or monitoring the other person; (b) isolate her (him) socially; (c) devalue her (him), denigrate her (him), humiliate her (him), or make her (him) feel bad about herself (himself); (d) cause others to turn against her (him), falsely accuse her (him); or blame her (him) for negative circumstances; (e) force her (him) to go against the law or her (him) moral and/or religious beliefs; or (f) destroy her (him) trust in herself or in the couple.

First, we select those items with the greatest variability in the sample. With this strategy, we pursued two objectives. First, by eliminating very unusual behaviors, multivariate kurtosis was reduced—an aspect that potentially affects the type of statistical analysis envisaged. Second, and related to the above, the grouping of some responses toward zero was also reduced, which, beyond reducing kurtosis, allows linear approximations in the estimated models. Initially, 19 items were selected from the CEPA’s PA subscales. These 19 items reflected the three conceptually relevant dimensions of PA, as discussed previously: control, humiliation, and coercion. Items from the originally PA subscale were selected if they met the following criteria: (a) they clearly reflected some of these dimensions (control, humiliation, and coercion), (b) they were not redundant with other items of the same subscale, and (c) they measured relatively frequent behaviors—that is, items that reflected very rare behaviors were excluded. At the end of this process, 19 items were finally selected. Separate principal component analysis for each dimension was conducted, and items with factor loadings lower than .50 were removed from the scale. At the end of this process, 14 items remained.

IMAFE

This is a self-describing personality questionnaire, according to 60 adjectives answered with a Likert-type scale of seven options (“never or almost never” to “always or almost always”; Lara, 1993). It evaluates four patterns of gender personality through the following 15 elements: (a) Masculinity, including adjectives relating to male and positive traits associated with practice and action; (b) Femininity, including positive feminine traits associated with relationships, social skills, mindfulness, and expressiveness; (c) Misogyny, including items with reference to features of aggression, domination, and intransigence; and (d) Submission, including feminine and negative traits related to self-sacrifice, dependence, subordination, and weakness.

The original validation of this instrument was based on data obtained with 1,301 Mexican men and women, forming a three-factor structure explaining 33.3% of the variance; the reliability indexes ranged between 0.74 and 0.92, and the scales correlated significantly with their counterparts in the Bem Sex Role Inventory (1974). Martínez-Gómez, Rodriguez Guerrero, and Rey-Anacona (2012) evaluated the reliability and structure validity of the instrument with data from 1,527 Colombian men and women aged between 15 and 42 years and found alpha values ranging between .76 and .88.

The factor analysis showed that the items were grouped in three factors explaining 30.8% of the variance, consistent with those found in the original cluster validation.

Procedures

Participants were selected by nonprobability sampling and were contacted directly in their classrooms in the educational institutions previously mentioned. Participants were informed about the duration of the implementation of the instruments, and their participation agreement was voluntary, anonymous, and confidential. In addition, they were informed that the information provided would not affect their educational process (only 10 students did not answer the instrument).

The participants answered the two instruments via self-report, whose were applied by the research team conducting the study, who had overseen the conditions in which the instruments had been implemented at physical and ethical levels. The team consisted of a master’s degree-level psychologist and a group of psychology students who had been previously trained in major conceptual and theoretical aspects of domestic violence and had been trained in employing the instrument through practical applications and behavioral test feedback.

Data Analysis

The EQS (Bentler, 1995) structural equations program, version 6.2, was used. The maximum likelihood estimator and Satorra–Bentler (S-B) χ² for correcting departure from multinormality were used for the calculation of the robust fit indexes (comparative fit index [CFI] and root mean square error of approximation [RMSEA]), standard errors, and statistical significance of the parameters. A CFI greater than .95 along with an RMSEA lower than .05 was considered a good fit for the model (Hooper, Coughlan, & Mullen, 2008). Similarly, the SPSS program (version 22.0) was used to examine the correlation structure of the instrument and to assess its overall reliability and scales through Cronbach’s alpha; additional comparisons were made by gender.

Results

Confirmatory factor analyses were performed with the 14 items of the Checklist. We first tested a single factor model in which all 14 items clustered into a unique latent variable—PA. This model fitted the data very poorly (S-B χ² = 807.67, df = 63, p < .001, CFI = .82, RMSEA = .12, 90% CI = [.11, .13]). These results suggested that a single first-order factor solution for the 14 items did not adequately represent the data (S-B χ² = 107.55, p < .001, df = 73, robust CFI = .95, robust RMSEA = .02, 95% CI = [.01, .02]). As this model was equivalent empirically to our theoretical model and found that model fit increased dramatically, we selected this second model as the best fitting model of CEPA in Adolescents and Young Adults. The three first-order latent factors were as follows: (a) Coercion, with four items; (b) Humiliation, with five items; and (c) Control, with five items (see Table 1). In this model, all the factor loadings were statistically significant (p < .001), and no cross loadings were allowed in the model. In addition, only one pair of correlated errors (within a first-order latent factor) was estimated by the model. The errors associated with the item (he or she controlled my time) and the item (he or she controlled my daily activities) of the factor “Control” were significantly correlated (r = .45, p < .001), suggesting that these two items shared common error variance. No further modifications were performed, as this final model showed an adequate fit to the data.

Table 1.

Items of the CEPA.

		Behavior	M	SD	Variance	α (If the Item Is Deleted)
Second-order factor (Psychological Abuse)	Factor I Coercion	42. He or she threatened to take you to a mental institution	0.04	0.267	.588	.801
		45. He or she threatened to kill you	0.05	0.259	.580	.786
		51 He or she threatened to kill a relative of her (him)	0.03	0.225	.705	.773
		69. He or she did not allow you to have contact with your family member	0.04	0.262	.428	.831
	Factor II Humiliation	28. He or she criticized your physical appearance (being fat, thin, etc.)	0.22	0.561	.483	.772
		58. He or she made you feel ridiculous, mocking of you in front of other people	0.15	0.470	.596	.725
		59. He or she said you were good-for-nothing	0.10	0.396	.593	.732
		64. He or she compared you negatively with other (women/men)	0.19	0.514	.590	.727
		77. He or she gave you nicknames to make you feel bad	0.11	0.410	.557	.741
	Factor III Control	1. He or she watched you in your place of study, work, or another different space	0.30	0.656	.337	.719
		35. He or she controlled your time	0.38	0.745	.589	.610
		38. He or she controlled your daily activities	0.29	0.682	.587	.611
		63. He or she searched in your handbag briefcase or your drawers without your consent	0.22	0.573	.495	.656
		82. He or she listened to your telephone conversation without your consent	0.11	0.403	.388	.700

Note. CEPA = Checklist of Experiences of Psychological Abuse to the Couple.

Unstandardized parameter estimates for the final model are presented in Table 2. All items loaded statistically significantly in their corresponding first-order factors, as seen by the large absolute z values (zs ≥ 5.96, p < .001) associated with each unstandardized factor loading. Second-order unstandardized factor loadings are also presented in the bottom part of Table 2, indicating that a second-order structure was adequately supported by the data, with second-order unstandardized factor loadings being highly significant (zs ≥ 4.28, ps < .001).

Table 2.

Unstandardized Estimates and Probability Associated for Final Model.

Items	Factor I(Coercion)	Factor II(Humiliation)	Factor III(Control)	Zs
He or she threatened to take you to a mental institution	1^a
He or she threatened to kill you	0.94			10.57
He or she threatened to kill a relative of her (him)	0.86			8.77
He or she did not allow you to have contact with your family member	0.87			8.16
He or she criticized your physical appearance (e.g., being fat, thin, etc.)		1^a
He or she made you feel ridiculous, mocking of you in front of other people		1.41		8.86
He or she said you were good-for-nothing		0.91		8.11
He or she compared you negatively with other (women) (men)		1.16		9.10
He or she gave you nicknames to make you feel bad		0.87		8.18
He or she watched you in your place of study, work, or another different space			1^a
He or she controlled your daily activities.			1.57	6.79
He or she searched in your handbag briefcase or your drawers without your consent			1.62	7.65
He or she listened to your telephone conversation without your consent			0.91	5.96
He or she controlled your time			1.70	7.34
Second-order factor	1	1.58	0.77	>4.28
Explained variance	56%	90%	47%

Note. All significant at p < .001.

Fixed to 1 during estimation.

In Table 3, we present the univariate descriptive statistics of the scale scores (mean and standard deviations) corresponding to the first-order factors (coercion, humiliations, and control) as well as to the second-order factor (psychological aggression). Reliability analyses and scores comparisons across gender were performed at this step, and the results are also presented in Table 3. Results from reliability analyses revealed that both the global score and the scale scores presented adequate indexes of internal consistency (.71 < αs < .84). The mean comparisons of scale scores across groups of gender did not find statistically significant differences between men and women. The exception was the Humiliation scale; women (M = 0.88, SD = 1.87) scored significantly higher than men (M = 0.71, SD = 1.57; p = .046).

Table 3.

Differences by Sex and Coefficient Reliability of CEPA.

Factor		n	M	SD	F	p	α
Control (five items)	Woman	824	1.40	2.23	2.93	.087	.712
	Man	673	1.21	1.98
	General	1,497	1.31	2.12
Humiliation (five items)	Woman	825	0.70	1.56	3.29	.070	.780
	Man	672	0.86	1.92
	General	1,497	0.77	1.73
Coercion (four items)	Woman	825	0.14	0.83	1.65	.200	.840
	Man	675	0.19	0.85
	General	1,500	0.16	0.835
Psychological Abuse (14 items)	Woman	818	2.23	3.71	0.00	.960	.835
	Man	669	2.24	3.99
	General	1,487	2.24	3.84

In Table 4, we present the bivariate correlations among the global score and scale scores of the Checklist and the IMAFE (Lara, 1993) scale scores. Interestingly, neither the masculinity scores nor the submission scores were statistically related to the psychological aggression global and scale scores. Although a pattern of significant correlations was found for the two constructs, these correlations, however, were in most cases in the low range. The global scale (Psychological Abuse) correlated positively and statistically with the three subscales, p ≤ .001, showing high correlations with Control and Humiliation and a more moderate correlation with Coercion.

Table 4.

Correlation Between IMAFE Scale (Lara, 1993) and CEPA (Pearson’s Correlation Coefficient).

Scales	M	F	Ma	S	Con	Hum	Coer	MP
MCorrelationSignificance (bilateral)	—
FCorrelationSignificance (bilateral)	.38** .000	—
MaCorrelationSignificance (bilateral)	.27 ** .000	−.24 ** .000	—
SCorrelationSignificance (bilateral)	−.27 ** .000	.04.255	.21 ** .000	—
ConCorrelationSignificance (bilateral)	−.02.584	−.07 * .048	.120 ** .001	.02.471	—
HumCorrelationSignificance (bilateral)	.04.232	−.122 ** .000	.17 ** .000	.02.578	.51 ** .000	—
CoerCorrelationSignificance (bilateral)	.00.982	−.09 ** .008	.04.202	.02.541	.35 ** .000	.54 ** .000	—
PACorrelationSignificance (bilateral)	.01.867	−.11 ** .002	.16 ** .000	.03.385	.86 ** .000	.85 ** .000	.65 ** .000	—

Note. IMAFE = Masculinity and Femininity Inventory; M = Masculinity; F = Feminism; Ma = Male chauvinism; S = Submission; Con = Control; Hum = Humiliation; Coer = Coercion; PA= Psychological Abuse.

Significant at .05 (bilateral). **Significant at .01 (bilateral).

Those results are coherent with the initial expectation because the hypothesis was that the gender negative behaviors such as aggression, independence, and dominance would be co-related in a positive form with the PA construct (control, humiliation, and coaction).

The three subscales also positively and significantly correlated with each other, p ≤ .001 level, but the correlations were moderate.

Discussion

The objective of this research was to evaluate the psychometric properties of the CEPA in a sample of Colombian adolescents and young adults, men and women. The results showed a structure composed for a factor of a higher order (“Psychological Aggression”), with 14 items, and three primary factors: (a) “Coercion” (four items), (b) “Humiliation” (seven items), and (c) “Control” (five items), with alpha indexes ranging from .71 to .83. In addition, the global scale and the three subscales correlated significantly with each other as well as personality traits of aggression, domination, and intransigence of IMAFE. Adjustment indexes were higher than 0.90, and the error was less than .05.

These results suggest that PA in young couples is a multidimensional phenomenon that can be subdivided into at least the following three subtypes: coercion (cognitive-behavioral), humiliation (emotional), and control (contextual-situational), which explains the high correlation that was observed between the general factor and each of these three sub-factors as well as moderate but statistically significant correlations that occurred between them. The PA factor explains most of the variance as well as the high internal consistency of the total scale and subscales, confirming that the items measure the same construct, which can be decomposed into the three sub-factors mentioned, in line with a psychosocial perspective study proposed by Rodríguez-Carballeira et al. (2005). Equally, the definition of PA as a multidimensional structure had already been reported by Rodríguez-Franco et al. (2010) through the validation of CUVINO, which defined eight factors to measure dating violence. Six factors were of a psychological nature, one factor was sexual violence, and the other factor was of physical violence.

This finding is also consistent with previous studies showing that PA is similarly exercised by men and women, because no statistically significant sex differences were found in the frequency of psychological abusive behaviors included in the three subscales and the full scale (e.g., Cortés-Ayala et al., 2015; Farhat et al., 2015).

As expected, the PA seems to both precede and accompany physical violence, justifying guides to intervene before such violent activity. The two subscales and the overall scale correlated significantly and positively with the scale of machismo of IMAFE (Lara, 1993), which contains self-descriptive traits of aggression, domination, and intransigence, and the three subscales along with the global scale correlated negatively and significantly with the scale of femininity, comprising features associated with interpersonal skills, concern for others, and expressiveness (see also Ramiro-Sánchez, Ramiro, Bermúdez, & Buela-Casal, 2018, for an analysis of dating violence and sexism). These results indicated, together with those mentioned above, that the CEPA has good construct validity.

Thus, the results obtained indicated that the CEPA has adequate psychometric properties and could be used in clinical practice and in the context of a comprehensive multi-methodical evaluation and research on violence in young couples, focusing on psychological violence. Psychological violence is the most common type of violence between couples of teenagers and young adults, as indicated by the prevalence studies (e.g., Almendros et al., 2009; Cortés-Ayala et al., 2015; Rey-Anacona, 2013; Sears et al., 2007). In addition, the resulting instrument can be applied faster than the original, which may be favorable in terms of research and clinical practice.

However, we are aware that this research suffers from significant limitations. It is worth noting both the conceptual heterogeneity and the various approaches that have been used in this evaluation of PA, and it must be emphasized that in some way, the intensity or severity of the PA, the frequency with which PA occurs, and/or chronicity of exposure to PA should be addressed. Future research should somehow incorporate the victim’s interpretation of the abusive act, the context, reciprocity, and/or the possible impact (Herrero, Vivas, Torres, & Rodríguez, 2018; Rodríguez-Franco et al., 2012). From this caution, further studies with the instrument are recommended, because the majority of the participants were single high school and college students geographically located in a region of Colombia. In addition, sampling was by availability, making it necessary to investigate the psychometric qualities of the instrument with randomized samples of adolescents and young adults with less education and other regions, as well as examining other types of reliability and validity that not were studied, such as the test–retest reliability and the criterion validity. It is also desirable to replicate this study with middle and older adults. Finally, although the results of the study could be generalized to partners of different sex (Martín-Fernández, Gracia, & Lila, 2018), future research should aim to verify the qualities of the instrument in other groups (e.g., same-sex couples, transgender, etc.). Thus, research in this field has traditionally been directed to the study of violence among heterosexual adolescents, while groups such as the LGTB (lesbians, gays, transsexuals and bisexuals) have traditionally been neglected (Santisteban & Gámez-Guadix, 2017).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

César Armando Rey Anacona, psychologist, PhD in Clinical and Health Psychology, is an associate professor of Pedagogical and Technological University of Colombia. César is also a senior researcher in Colciencias, Colombia.

Jorge Arturo Martínez Gómez, psychologist, magister in Clinical Psychology, PhD student in Psychology, San Buenaventura University Medellín-Colombia. Jorge is a professor of Pedagogical and Technological University of Colombia and also a junior researcher in Colciencias, Colombia.

Juan B. Herrero Olaizola, PhD in Psychology from the University of Valencia (Spain), is a professor of Social Psychology at the University of Oviedo, researcher and director of the Group of Community, Legal and Health Psychology (CJS) of the University of Oviedo (Spain).

Francisco Javier Rodriguez Díaz, PhD in Psychology from the University of Santiago de Compostela (Spain), is a professor of Personality, Evaluation and Psycho-logical Treatment at the University of Oviedo. Researcher of the Group of Community, Legal and Health Psychology (CJS) of the University of Oviedo (Spain).

References

Almendros

Gámez-Guadix

Carrobles

J. A.

Rodríguez-Carballeira

Porrúa

(2009). Intimate partner psychological abuse: Concept, measurement, and recent contributions. Behavioral Psychology/Psicología Conductual, 17, 433-451.

Bentler

P. M.

(1995). EQS structural equations program manual. Encino, CA: Multivariate Software.

Bringas-Molleda

Cortés-Ayala

Flores-Galaz

Antuña Bellerín

M. A.

López-Cepero Borrego

Rodríguez-Díaz

F. J.

(2015). Análisis diferencial de la percepción de jóvenes sobre maltrato en el noviazgo [Differential analysis of the perception of abuse in relationships among young]. Revista Latinoamericana de Ciencias Sociales, Niñez y Juventud, 13, 477-488.

Browne

Herbert

(1997). Preventing family violence. Rochester, NY: John Wiley.

Centers for Disease Control and Prevention. (2006). Physical dating violence among high school students, United States, 2003. Morbidity and Mortality Weekly Report, 55, 532-535.

Colombian Institute of Legal Medicine and Forensic. (2005). Forensis: Datos para la vida. Bogotá, Colombia.

Cortés-Ayala

Bringas-Molleda

C. B.

Rodríguez-Franco

Flores Galaz

Ramiro-Sánchez

Rodríguez-Díaz

F. J.

(2014). Unperceived dating violence among Mexican students. International Journal of Clinical and Health Psychology, 14, 39-47. doi:10.1016/S1697-2600(14)70035-3

Cortés-Ayala

Flores Galaz

Bringas-Molleda

Rodríguez-Franco

López-Cepero Borrego

Rodríguez-Díaz

F. J.

(2015). Relación de maltrato en el noviazgo de jóvenes mexicanos: Análisis diferencial por sexo y nivel de estudios [Intimate partner violence in the relationships of mexican youth. Differential analysis by sex and level of schooling]. Terapia Psicológica, 33, 5-12.

Delgado-Alavarez

Estrada-Aranda

López-Huerto

J. A.

(2015). Gender and cultural effects on perception of psychological violence in the partner. Psicothema, 27, 381-387.

10.

Evers

Muñiz

Hagemeister

Høstmælingen

Lindley

Sjöberg

Bartram

(2013). Assessing the quality of tests: Revision of the EFPA review model. Psicothema, 25, 283-291.

11.

Exner-Cortens

Gill

Eckenrode

(2016a). Measurement of adolescent dating violence: A comprehensive review (Part 1, behaviors). Aggression and Violent Behavior, 27, 64-78. doi:10.1016/j.avb.2016.02.007

12.

Exner-Cortens

Gill

Eckenrode

(2016b). Measurement of adolescent dating violence: A comprehensive review (Part 2, attitudes). Aggression and Violent Behavior, 27, 93-106. doi:10.1016/j.avb.2016.02.011

13.

Farhat

Haynie

Summersett-Ringgold

Brooks-Russell

Iannotti

R. J.

(2015). Weight perceptions, misperceptions, and dating violence victimization among U.S. adolescents. Journal of Interpersonal Violence, 30, 1511-1532. doi:10.1177/0886260514540804

14.

Fernández-Fuertes

Fuertes

Pulido

(2006). Evaluación de la violencia en las relaciones de pareja de los adolescentes. [Validación del Conflict in Adolescent Dating Relationships Inventory (CADRI) - versión española]. International Journal of Clinical and Health Psychology, 6(2), 339-358.

15.

Garrido

E. F.

Taussig

H. N.

(2013). Do parenting practices and prosocial peers moderate the association between intimate partner violence exposure and teen dating violence? Psychology of Violence, 3, 354-366. doi:10.1037/a0034036

16.

Overbeek

Cillessen

A. H. N.

Engels

R. C. M. E

. (2012). A longitudinal study of the associations among adolescent conflict resolution styles, depressive symptoms, and romantic relationship longevity. Journal of Adolescence, 35, 1247-1254. doi:10.1016/j.adolescence.2012.04.009

17.

Herrero

Rodríguez-Díaz

F. J.

Torres

(2017). Acceptability of partner violence in 51 societies: The role of sexism and attitudes toward violence in social relationships. Violence Against Women, 23, 351-367.

18.

Herrero

Torres

Rodríguez

F. J.

(2018). Child abuse, risk in male partner selection, and intimate partner violence victimization of women of the European Union. Prevention Science, 19, 1102-1112. doi:10.1007/s11121-018-0911-8

19.

Herrero

Torres

Rodríguez

F. J.

Juarros-Basterretxea

(2017). Intimate partner violence against women in the European Union: The influence of male partners’ traditional gender roles and general violence. Psychology of Violence, 7, 385-394.

20.

Herrero

Vivas

Torres

Rodríguez

F. J.

(2018). When violence can appear with different male partners: Identification of resilient and non-resilient women in the European Union. Frontiers in Psychology, 9, 877.

21.

Hokoda

Ramos-Lira

Celaya

Vilhauer

Angeles

Ruíz

. . . Duque

(2006). Reliability of translated measures assessing dating violence among Mexican adolescents. Violence and Victims, 21, 117-127.

22.

Hooper

Coughlan

Mullen

(2008). Structural equation modelling: Guidelines for determining model fit. Electronic Journal of Business Research Methods, 6, 53-60.

23.

Lara

M. A.

(1993). Inventario de Masculinidad y Feminidad [Inventory of Masculinity and Femininity México]. México: Manual Moderno.

24.

Lila

Gracia

Murgui

(2013). Psychological adjustment and victim-blaming among intimate partner violence offenders: The role of social support and stressful life events. The European Journal of Psychology Applied to Legal Context, 5, 73-95.

25.

López-Cepero Borrego

Rodríguez-Franco

Rodríguez-Díaz

F. J

. (2015). Evaluación de la violencia de pareja. Una revisión de instrumentos de evaluación conductual [Measuring intimate partner abuse. A review of behavioral assessment]. Revista Iberoamericana de Diagnóstico y Evaluación Psicológica, 40, 37-50.

26.

López-Cepero Borrego

Rodríguez-Franco

Rodríguez-Díaz

F. J.

Bringas-Molleda

Paino

S. G

. (2015). Percepción de la victimización en el noviazgo de adolescentes y jóvenes españoles. [Perception of victimization in dating in Spanish teen and young]. Revista Iberoamericana de Psicología y Salud, 6, 64-71.

27.

Martín-Fernández

Gracia

Lila

(2018). Assessing victim-blaming attitudes in cases of intimate partner violence against women: Development and validation of the VB-IPVAW scale. Psychosocial Intervention, 27, 133-143.

28.

Martínez-Gómez

J. A.

Guerrero-Rodríguez

S. I.

Rey-Anacona

C. A.

(2012). Evaluación de la validez de constructo y la confiabilidad del Inventario de Masculinidad y Femineidad en adolescentes y adultos jóvenes colombianos. Avances en Psicología Latinoamericana, 30(1), 170-181.

29.

Messinger

A. M.

Rickert

V. I.

Fry

D. A.

Lessel

Davidson

L. L.

(2012). Revisiting the role of communication in adolescent intimate partner violence. Journal of Interpersonal Violence, 27, 2920-2935. doi:10.1177/0886260512438276

30.

Ministry of Health of Colombia. (1999). Norma para el diagnóstico y atención integral de mujer maltratada. Bogotá, Colombia: Author.

31.

Muñoz-Rivas

M. J.

Andreu-Rodríguez

J. M.

Graña-Gómez

J. L.

O’Leary

D. K.

González

M. P.

(2007). Validación de la versión modificada de la Conflicts Tactics Scale (M-CTS) en población juvenil española [Validation of the modified version of the Conflict Tactics Scale (M-CTS) in a Spanish population of youths]. Psicothema, 19, 693-698.

32.

Osorio-Guzmán

Tani

Bazán

R. G.

Bonechi

Menna

(2012). Cuestionario Maltrato en el Noviazgo (CMN): Instrumento Binacional (Italia-México) [Abuse in dating questionnaire (CMN): Binational instrument. (Italy-Mexico)]. Revista de Psicología, 14, 47-60.

33.

Ramiro-Sánchez

Ramiro

M. T.

Bermúdez

M. P.

Buela-Casal

(2018). Sexism in adolescent relationships: A systematic review. Psychosocial Intervention, 27, 123-132. doi:10.5093/pi2018a19

34.

Reidy

D. E.

Smith-Darden

Cortina

K. S.

Kernsmith

R. M.

Kernsmith

P. D.

(2015). Masculine discrepancy stress, teen dating violence, and sexual violence perpetration among adolescent boys. Journal of Adolescent Health, 56, 619-624. doi:10.1016/j.jadohealth.2015.02.009

35.

Rey-Anacona

C. A.

(2009). Características psicológicas de adolescentes varones y adolescentes mujeres victimarios de violencia en su relación de noviazgo [Psychological characteristics of adolescent males and adolescent female victims of violence in their relationship of courtship] [Research report]. Tunja: Pedagogical and Technological University of Colombia.

36.

Rey-Anacona

C. A.

(2013). Prevalencia y tipos de maltrato en el noviazgo en adolescentes y adultos jóvenes [Prevalence and types of dating violence in adolescents and young adults]. Terapia Psicológica, 31, 143-154.

37.

Rey-Anacona

C. A.

(2015). Variables asociadas a los malos tratos en el noviazgo en adolescentes y adultos jóvenes [Variables associated with dating violence in adolescents and young adults]. Acta Colombiana de Psicología, 18, 159-171.

38.

Rey-Anacona

C. A.

Mateus

A. M.

Bayona

P. A.

(2010). Malos tratos ejercidos por adolescentes durante el noviazgo: Diferencias por sexo [Mistreatment by adolescents while dating: Differences by sex]. Revista Mexicana de Psicología, 27, 169-181.

39.

Rodríguez-Carballeira

Almendros

Escartín

Porrúa

Martín-Peña

Javaloy

Carrobles

J. A.

(2005). Un estudio comparativo de las estrategias de abuso psicológico: en pareja, en el lugar de trabajo y en grupos manipulativos [A comparative study of psychological abuse strategies: a couple, in the workplace and manipulative groups]. Anuario de Psicología, 36, 299-314.

40.

Rodríguez-Díaz

F. J.

Herrero

Rodríguez-Franco

Bringas-Molleda

Paíno-Quesada

S. G.

Pérez

(2017). Validation of Dating Violence Questionnaire-R (DVQ-R). International Journal of Clinical and Health Psychology, 17, 77-84.

41.

Rodríguez-Franco

López-Cepero Borrego

Rodríguez-Díaz

F. J.

(2009). Violencia doméstica: Una revisión bibliográfica y bibliométrica [Domestic violence: A bibliographic and bibliometric review]. Psicothema, 21, 253-259.

42.

Rodríguez-Franco

López-Cepero Borrego

Rodríguez-Díaz

F. J.

Bringas-Molleda

Antuña

Estrada

(2010). Validation of the Dating Violence Questionaire, DVQ (CUVINO) among Spanish-Speaking younth: Analysis of resultd in Spain, Mexico and Argentina. Anuario de Psicología Clínica y de la Salud, 6, 45-52.

43.

Rodríguez-Franco

López-Cepero Borrego

Rodríguez-Díaz

F. J.

Bringas-Molleda

Estrada Pineda

Antuña

M. A.

Quevedo-Blasco

(2012). Labeling dating abuse: Undetected abuse among Spanish adolescents and young adults. International Journal of Clinical and Health Psychology, 12, 55-67.

44.

Rubio-Garay

López-González

M. A.

Carrasco

M. A.

Amor

P. J.

(2017). Prevalencia de la violencia en el noviazgo: una revisión sistemática [The prevalence of dating violence: a systematic review]. Papeles del Psicólogo, 38, 135-147. doi:10.23923/pap.psicol2017.2831

45.

Santisteban

P. D.

Gámez-Guadix

(2017). Estrategias de persuasión en grooming online de menores: un análisis cualitativo con agresores en prisión [Strategies of persuasion in the online hygiene of minors: a qualitative analysis with aggressors in prison]. Psychosocial Intervention, 26, 139-146.

46.

Sears

H. A.

Byers

E. S.

Price

E. L.

(2007). The co-occurrence of adolescent boys’ and girls’ use of psychologically, physically, and sexually abusive behaviours in their dating relationships. Journal of Adolescence, 30, 487-504.

47.

Shorey

R. C.

Seavey

A. E.

Brasfield

Febres

Fite

P. J.

Stuart

G. L.

(2015). The moderating effect of social support from a dating partner on the association between dating violence victimization and adjustment. Violence Against Women, 21, 460-477. doi:10.1177/1077801215570482

48.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The Conflict Tactics Scales (CTS). Journal of Marriage and Family, 41, 75-88.

49.

Straus

M. A.

Douglas

E. M.

(2004). A short form of the Revised Conflict Tactics Scales, and typologies for severity and mutuality. Violence and Victims, 19, 507-520. doi:10.1891/088667004780927800

50.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The Revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.

51.

Taylor

Calkins

C. A.

Xia

Dalla

R. L.

(2017). Adolescent perceptions of dating violence: A qualitative study. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517726969

52.

Vizcarra Larrañaga

M. B.

Póo

A. M

. (2011). Violencia de pareja en estudiantes universitarios del sur de Chile. [Intimate Partner Violence in South Chilean College Students]. Universitas Psychologica, 10, 89-98.

53.

Williams

Miller

Cutbush

Gibbs

Clinton-Sherrod

Jones

(2015). A latent transition model of the effects of a teen dating violence prevention initiative. Journal of Adolescent Health, 56(2), S27-S32. doi:10.1016/j.jadohealth.2014.08.019

54.

Wolfe

D. A.

Scott

Reitzel-Jaffe

Wekerle

Grasley

Straatman

A. L.

(2001). Development and validation of the Conflict in Adolescent Dating Relationships Inventory. Psychological Assessment, 13, 277-293. doi:10.1037/1040-3590.13.2.277