1. Introduction

Intimate partner violence (IPV) is a phenomenon that is claiming more and more victims, while studies show that the relationship patterns of the members of the couple al ready appear during dating relationships, constituting a risk factor for suffering IPV ^{(Amor et al., 2001}; ^{García et al., 2018}; ^{García-Carpintero-Muñoz et al., 2022}; ^{Gutiér rez et al., 2022}; ^{Rubio-Garay et al., 2012}; ^{Yanez-Peñúñuri et al., 2019)}. The Pan American Health Organization ^{(PAHO, 2021)} reports that one in three women have ex perienced or are experiencing physical or sexual violence from their partner or ex-partner and that three out of every five murders of women are committed by their partner.

Additionally, ^{Bott et al. (2021)} found that intimate partner violence is the most common form of violence against women in Latin America and ^{García Cabezas (2020)} also reports that a femicide is recorded. A system atic search of surveys on violence in ^{PAHO (2021)} Member States, with 1046 records from 24 countries between 1998 and 2018, shows that this type of violence is present across the continent, with prevalence ranging from 1% in Canada to 50% in Bolivia in the past 12 months.

Bolivia recognizes feminicide within the legislation in 2013, reporting up to 2021 a total of 998 femicides. ^{Choque Aldana (2021)} reports that Bolivia has third high est rate of femicide in Latin America and the Caribbean, and the National Institute of Statistics (INE) reported the presence of almost 52% of at least one episode of violence coming from their partner or ex-partner. Along these lines, it is worth highlighting the studies by ^{Rodríguez-Franco et al. (2022)}, with young Spaniards, and ^{Cortés Ayala et al. (2014)}, with young Mexicans, which report that more than 70% of the participants in their studies did not perceive themselves as mistreated by their part ner, despite the evidence that they had experienced abuse in their dating relationship. The authors assume that vi olence is probably not recognized due to the lack of in formation about it, its normalization or even because it is confused with displays of affection. Therefore, dating violence offers such disparate prevalence figures, ranging between 9% and 90% ^{(Rubio-Garay et al., 2017)}.

This leads ^{Escoto et al. (2007)} to consider that ado lescents create great expectations of love, care, support, and understanding in courting, which prevents them from noticing when they become involved in dating violence. This will also lead to a great diversity of definitions, ideolo gies, and methodological positions in the different groups and contexts in which it has been investigated. The im portance of studying this phenomenon lies in its predic tive nature of future aggression, something that will be come increasingly serious with time and coexistence ^{(Rey Anacona et al., 2010}; ^{Rey-Anacona et al., 2021)}.

It is understood that dating violence is any act, omis sion, attitude or expression that generates, or has the po tential to generate emotional, physical or sexual harm to the affective partner with whom an intimate relationship is maintained in the absence of cohabitation, economic bond, legal or marital status. It has specific character istics not shared with the violence experienced in other stages of the life of a couple ^{(Rubio-Garay et al., 2017)}, while as a social problem, from the legal perspective and linked to the intervention of power, it is postulated that it must imply intentionality, not being desired, not be ing essential, and being harmful ^{(Hamby, 2017)}, where it will be necessary to discriminate the perception of the behavior and if it is an act or appears as a process ^{(García-Díaz et al., 2020}; ^{López-Cepero et al., 2015)}. A characteristic of this type of violence is its bidirectional nature regardless of the sex of the partners, especially when it comes to psychological and physical violence. It has been seen that as there is an increase in the level of aggressiveness. This characteristic is lost and more women are affected as the relationship becomes more stable ^{(Herrero-Olaizola et al., 2020}; ^{Rey-Anacona et al, 2021}; ^{Riesgo et al., 2019}; ^{Rubio-Garay et al., 2017)}.

This approach, in addition to making its identifica tion more precise and improving its ‘surveillance’, will promote a better evaluation and may offer a better guide to its treatment ^{(Hamby, 2017)}. This implies the need to have effective instruments aimed at evaluating the presence of violence in couple relationships, either those aimed at diagnosing abuse within the couple/dating vio lence (long instruments) or those that seek to make a de tection orientation or screen. However, many of them lack methodological rigor in their validation process, stand ing out three for work with Spanish-speaking samples: Conflict Tactics Scale 2 (CTS2), Conflict in Adolescent Dating Relationships Inventory (CADRI), and the Dat ing Violence Questionnaire (DVQ) ^{(López-Cepero et al., 2015)}. The first refers to not being prepared for working with young people and adolescents, and the second to not delving into the study of victimization, in addition to preserving Anglo-Saxon elements that probably do not adapt to the context. The DVQ is proposed the most appropriate to study this topic; even more so with its re vised and reduced version, the DVQ-VP ^{(Rodríguez-Díaz et al., 2017}; ^{Rodríguez-Franco et al., 2022)}.

It is worth mentioning that most of the tests that have psychometric support have been prepared and val idated with samples from the United States or Europe, which could ignore the importance of the cultural com ponent and the sociodemographic context as determi nants of the phenomenon studied ^{(Fernández et al., 2010)}. In this sense, the original version of the DVQ-VP has been worked with the Latin American population, and currently has a version that also allows the study of the phenomenon from both the victim and the aggres sor ^{(Rodríguez-Franco et al., 2022)}. Thus, this study aims to confirm the hypothesis that the DVQ-VP fit a five-factor model that assesses dating violence in the Bolivian population.

2. Method

3. Results

4. Discussion

The present investigation tested the trichotomic model proposed by the authors of the scale in 2021, analyzing additionally the validity and reliability of the DVQ-RP ^{(Rodríguez-Franco et al., 2022)}. The exploratory factor analyzes allow us to appreciate that the items converge in two dimensions in the case of the questionnaire for vic timization and one dimension in the case of perpetrated violence. However, in the case of the DVQ-R, the ad justment values of the model confirm its configuration in five factors ^{(Rodríguez-Díaz et al., 2017}; ^{Rodríguez-Franco et al., 2022)} with a modification of the previous version: the elimination of items 4 and 13 correspond ing to the categories of detachment and coercion, respec tively. In the previous studies reviewed ^{(Rodríguez-Díaz et al., 2017}; ^{Rodríguez-Franco et al., 2022)}, these two dimensions report lower reliability values, which could be due to the mediation of culture and the situation in which it is considered or not considered that violence occurs. Similarly, it is worth mentioning that there are some differences between the textual expression of the original version in Spanish and the official translated ver sion, which in the Spanish-speaking context may have differences depending on the country (see Appendix 1). However, it should also be considered that the discrim inant validity criteria have not been contemplated in previous adaptations.

Although the presence of intention, harm, and that the conduct is not desired or necessary is taken into ac count for a conduct to be called violence ^{(Hamby, 2017)}, these criteria can be observable and clearly measurable when it comes to physical violence, sexual or insults that involve active and aggressive verbal conduct in which the intention of the issuer and the damage caused to the victim can be seen. However, as the behavior becomes non-verbal and the aggressiveness is attenuated or hid den in the case of subtle psychological violence ^{(Novo et al., 2016)}, the context can acquire greater weight ^{(Álvarez Roldán & Parra Toro, 2012)}, which, added to the fact of being in love, could decrease the perception of the other person’s control intentions ^{(Bartels & Zeki, 2004)}.

In this sense, given that we worked with young peo ple in a dating relationship and that romantic love is usually at its highest point, it could happen that vio lent actions such as emotional blackmail or disinterest are overlooked, and instead are interpreted as expres sions of affection, interest or parts of the personality of the loved one ^{(Cubells-Serra et al., 2021}; ^{Marcos et al., 2020}; ^{Standley, 2022)}. Thus, possibly the expression “Has physically kept you from leaving”, since it does not have a condition of use of force or shows the inten tion of the person issuing the behavior, it can pass as an expression of affection or a need to spend more time together. It is also the particular case of item 4 (“Is a good student, but is always late at meetings, does not fulfil his/her promises, and is irresponsible”) since, in addition to the above, it is possible that it is not consid ered as something violent within Bolivian society, due to a cultural factor called “the Bolivian hour”, where lateness is normalized and is not necessarily taken as a lack of respect or a aggression, but as something circum stantial and very common that, in many cases, will not depend on the person but on the situations in which they find themselves ^{(Águeda Cotjiri, 2017}; ^{Cosio Hurtado, 2010)}. In this regard, it could be interesting to study the behavior of this reagent in other countries and contexts.

In turn, the lack of discriminant validity and the high correlations between the various factors suggest that the different manifestations of violence in dating relation ships are interlinked, there being a second order factor since it would not be enough to understand the violence as physical (physical and sexual) and psychological (hu miliation, detachment, and coercion) or “active” and “passive” in a disjointed way. This was observed when trying to confirm the two-factor model resulting from the AFE. In this sense, it should be remembered that a person who is a victim of sexual or physical violence will also be experiencing the consequences of psychological violence since, although the violence does not follow lin ear steps, the aggressions usually worsen as can be seen graphically in the violence meter originally proposed by the National Polytechnic Institute of Mexico ^{(Tronco Rosas & Ocaña López, 2011)}. An important element to consider is the presence of invariance with respect to sex, both for the victimization questionnaire and for the exercise of violence, since it will provide more informa tion on the bidirectionality of the phenomenon and the need to study men and women in both roles.

It can be said that the DVQ-VP is a valid and reli able test, as it has been analyzed from its psychometric properties, both for experienced violence (DVQ-R) and for perpetrated violence (DVQ-RP).

5. Limitations

The perception of intentionality may differ in the inter pretation that one has of a behavior as violent or not, depending on whether the role of victim or aggressor is exercised. In this sense, one of the limitations of this study is the population with which it was worked: uni versity students with a relationship in the last twelve months, and their self-report being the only data avail able. Therefore, given the bidirectional nature of dating violence ^{(Herrero-Olaizola et al., 2020}; ^{Riesgo et al., 2019}; ^{Rojas-Solís & Romero-Méndez, 2022}; ^{Rubio-Garay et al., 2017}; ^{Rubio-Garay et al., 2012)}, working with both members of the couple simultaneously would be more fruitful and would allow a better evaluation of the role of intention behind the behaviors and their interpretation, depending on whether role of victim or aggressor, mainly in the case of violence by detachment and coercion. In addition, another factor that could be linked to the interpretation of one’s own behavior as violent or not, could be its justification, based on mecha nisms of moral disconnection ^{(Rubio-Garay et al., 2019)}, mediating along with the level of detachment regarding the intensity and frequency of the violence, as we tried to improve the adaptation of the items by introducing the use of the radio or letters in the reagents, as well as thinking of the digital era that currently prevails by changing the radio by music (currently there can be var ious devices linked to this activity) and letters to all instant messaging and any formal means (such as email or social networks).

6. Conclusion

The appropriate psychometric characteristics were con firmed for evaluating violence in dating relationships us ing a reduced version of 18 items in a five-factor model for victims and a single-factor scale for the evaluation of aggressors, which were found to be invariant for sex. Be ing a self-report, it is necessary in future studies to com pare the information with that provided by the couple.