The First Comprehensive Assessment of Offline and Online Emotional Intimate Partner Violence (EIPV) in a Representative Sample of Young Adults in Poland

Abstract

The current study provides a comprehensive assessment of emotional intimate partner violence (EIPV) among a representative sample of young adults (aged 18–29 years) in Poland (N = 2,705; 53.4% female). We found that 80% of the sample were victimized and perpetrated offline EIPV more than twice in their lifetime. Online EIPV affected 40% of participants more than twice in the lifetime. Offline and online EIPV as well as EIPV victimization and perpetration behaviors tended to co-occur. The perpetrator–victim phenomenon was confirmed by the results of the latent profile analysis (LPA), which uncovered three latent groups characterized by perpetration-victimization co-occurrence.

Keywords

emotional intimate partner violence (EIPV) victimization emotional intimate partner violence (EIPV) perpetration perpetrator–victim phenomenon young adults latent profile analysis (LPA)

Introduction

Emotional intimate partner violence (EIPV) is the most common form of partner abuse, with young adults being at the greatest risk of both victimization and perpetration (Dokkedahl et al., 2019; Doroszewicz & Forbes, 2008; World Health Organization [WHO], 2013). EIPV, also referred to as psychological IPV, encompasses behaviors such as denigration, dominance and intimidation, coercive control, threats to physical and psychological health, and isolation (Dokkedahl et al., 2022). Some research demonstrated that women and men abuse their partners at similar rates (Hines & Douglas, 2009). When women commit IPV, it is predominantly in the form of psychological abuse (Williams et al., 2008). However, to date, most studies have focused on males as perpetrators and physical forms of IPV. Risk factors and consequences associated with IPV include rural residence, low education level, social isolation, being young, psychiatric disorder, history of victimization, low self-esteem, higher loneliness, and increased mental distress (e.g., Barnes et al., 2022; Okuda et al., 2015; Ortega Pacheco et al., 2024; Reichel, 2017; Spencer et al., 2024). Although some studies indicate that EIPV victimization can sometimes have a stronger effect on health outcomes than physical IPV (Pico-Alfonso, 2005), most prior research considered total IPV experiences or its physical and/or sexual forms, leaving EIPV blatantly understudied. In spite of its prevalence and serious consequences, EIPV remains the least understood and studied form of IPV (Dokkedahl et al., 2022). Therefore, the main aim of the current study was to provide a comprehensive assessment of EIPV victimization and perpetration, including prevalence rates and patterns of co-occurrence between different EIPV behaviors, among a representative sample of young adults in Poland.

Approximately 50% of men and women experience EIPV, making it the most common form of IPV (Spencer et al., 2024). Cyber partner abuse, also referred to as online EIPV, received less research attention, but one study estimated that 40% of American college students were victimized in this way (Wolford-Clevenger et al., 2016). Cyber partner abuse becomes increasingly widespread, especially in the post-pandemic world where many social interactions, including those within romantic relationships, take place online. Research shows that offline and online IPV tend to be strongly correlated (Gilbar et al., 2023; Schokkenbroek et al., 2022), indicating that the same individuals are likely to experience IPV across the two settings. Nevertheless, reports of IPV are subject to social desirability bias and so IPV tends to be underreported (Wenger, 2012). The underreporting of EIPV may also be due to the lack of recognition of EIPV as abuse. Indeed, studies show that individuals who experience EIPV may not acknowledge these behaviors as a source of stress and some acts of psychological abuse may even appear affectionate (Minto et al., 2022). EIPV is also likely to be bidirectional, that is, the same individual receives and uses abusive behaviors (O’Leary & Slep, 2012; Wagers et al., 2021). Prior research revealed moderate to strong correlations between EIPV victimization and perpetration in offline and online settings (Schokkenbroek et al., 2022). A recent study into dating violence among Spanish university students showed that the reciprocity of IPV was stronger for offline abuse (Jaureguizar et al., 2024). In a rare study into bidirectional IPV, Ulloa and Hammett (2016) found that perpetration-victimization was associated with more negative mental health outcomes than no violence, perpetration-only, and victimization-only.

As for the case of Poland, statistical reports on IPV rates are rare and provide inconsistent data. For example, in a review of studies among ever-partnered women aged between 15–49 years in 161 countries, Poland was classified among the countries with some of the lowest rates of lifetime physical and sexual violence (13%) (Sardinha et al., 2022). However, two other studies indicate that approximately 30% of community-based women in Poland experienced IPV in the last 12 months (Debowska et al., 2025; Łukasik et al., 2015). In a study by Doroszewicz and Forbes (2008) conducted among female and male college students in Poland (N = 201; 49.8% female), the rates of physical violence and sexual coercion were also high: 36% and 42% for men and 48% and 40% for women respectively. In the same study, EIPV was recognized as the most common type of dating violence, affecting 77% of men and 89% of women. However, to the best of our knowledge, there are no published reports on EIPV prevalence in the Polish population and very little is known about relationship violence in this part of Europe, especially in the context of EIPV victimization among men. Although more comprehensive statistics are not available, it is believed that Poland, a predominantly Catholic country with strong bonds to traditions, has a high prevalence of IPV and domestic violence and that there is a widespread acceptance of this form of violence in the Polish society (Brunell, 2005; Doroszewicz & Forbes, 2008).

Noteworthy, individuals with a lived experience of IPV do not constitute a homogenous group. As mentioned above, some people are both victimized and perpetrate IPV (the perpetrator–victim phenomenon), some people perpetrate a specific form of violence, and some others engage in multiple violence perpetration. In order to uncover how various forms of IPV co-occur with one another, scholars have utilized person-centered analytic approaches, such as latent class analysis (LCA) and latent profile analysis (LPA). The results of Hall et al.'s (2023) recent systematic review of IPV victimization profiles indicated that the most commonly retrieved IPV victimization groups included: no/low violence, physical and psychological victimization, and multiple victimization. As for associated consequences, results demonstrated that as the abuse increased, so did the negative outcomes. To date, only one LCA study profiled IPV experience using both victimization and perpetration variables. The study, conducted by Grest et al. (2018) among Latino adolescents from Southern California (N = 1,060; 60.6% female), retrieved four latent classes, with the largest one being characterized by no IPV victimization and perpetration experiences (78.1% of the sample). Interestingly, all three remaining classes were characterized by bidirectional IPV victimization and perpetration: a sexual IPV class (11.3%), a psychological IPV class (8.1%), and a multi-form IPV class (2.5%). However, although IPV is experienced by both sexes (Breiding, 2015), most prior LCA/LPA studies focused on women, leaving an important knowledge gap. Research that included female and male participants revealed that women were more likely to be classified in groups characterized by more severe abuse victimization (Ansara & Hindin, 2010; Carbone-López et al., 2006; Lagdon et al., 2023). According to Hall et al. (2023), EIPV is the most omitted form of IPV in LCA/LPA research—even though it is consistently documented as the most prevalent form of abuse (Dokkedahl et al., 2019)—and no profiling studies to date focused exclusively on EIPV behaviors. A better understanding of the nature of emotionally abusive relationships is vital for more efficient preventive efforts and to promote healthy romantic relationships.

The Present Study Objectives and Predictions

The current study offers the first comprehensive assessment of EIPV among a representative sample of Poles and has been designed with the above-mentioned literature gaps in mind. The overarching aim of the current study is to build a better understanding of the prevalence, patterns of co-occurrence, risk factors, consequences, and recognition of offline and online EIPV among young adults (aged 18–29 years) in Poland. For the purpose of this study, offline EIPV is defined as psychological abuse occurring in-person, characterized by behaviors across four main dimensions: restrictive engulfment (RE; efforts to track, monitor, and isolate a partner), denigration (humiliating and belittling), hostile withdrawal (HW; withholding communication or attention coercively), and dominance/intimidation (DI; using threats or destruction of property to achieve compliance). Conversely, online EIPV (or cyber partner abuse) refers to psychological abuse perpetrated via technology and digital mediums. This encompasses direct aggression, such as using technology to send threats or insults intended to hurt the partner, as well as monitoring and control, which involves using technology to digitally track and control a partner's activities.

The study examines the following main research objectives:

To establish the past-year and lifetime prevalence of offline and online EIPV victimization and perpetration. We predict that at least 1 in 2 men and women in the current sample will report lifetime EIPV victimization and perpetration experiences (Spencer et al., 2024). Prevalence rates will be similar for female and male participants, as prior research demonstrates that men and women perpetrate psychological abuse at comparable rates (Hines & Douglas, 2009).

To determine whether there is a significant correlation between online and offline EIPV victimization/perpetration. We predict moderate to strong positive correlations between EIPV offline and online. This prediction is drawn from recent studies indicating that offline and online IPV tend to be strongly correlated, suggesting the same individuals experience IPV across both digital and physical settings (Gilbar et al., 2023; Schokkenbroek et al., 2022).

To determine whether there is a significant correlation between EIPV perpetration and victimization. Because EIPV is highly likely to be bidirectional (O’Leary & Slep, 2012; Wagers et al., 2021), we predict moderate to strong positive correlations between EIPV perpetration and victimization that take place in the same setting (i.e., offline or online) and moderate positive correlations between EIPV perpetration and victimization that take place in a different setting. This distinction is informed by research showing that IPV reciprocity is generally stronger for abuse occurring within the same medium, particularly offline (Jaureguizar et al., 2024; Schokkenbroek et al., 2022).

To determine whether women and men who experience/perpetrate EIPV recognize the harm they are experiencing/inflicting. We predict that there will be a weak positive association between EIPV victimization and recognition of experienced harm among both sexes. Analogously, there will be a weak positive association between EIPV perpetration and recognition of inflicted harm. These predictions are based on findings that individuals often fail to recognize psychologically abusive behaviors as a source of stress or harm, sometimes even misinterpreting them as affectionate (Minto et al., 2022).

To establish profiles of offline and online EIPV victimization and perpetration and associated external criteria (sociodemographic risk factors and consequences): sex, age, level of education (higher/non-higher), place of residence (urban/rural), sexual orientation (heterosexual/non-heterosexual), history of child abuse and neglect (CAN), psychiatric diagnosis (present/absent), global health status, current mental distress, self-esteem, emotional loneliness, and social loneliness. Given the lack of past EIPV profiling studies, we did not formulate any a priori hypotheses regarding the number of EIPV profiles. However, extrapolating from prior latent class research on IPV (Grest et al., 2018; Hall et al., 2023), we expect to find the following groups: a group characterized by low scores on all EIPV variables, a group characterized by increased EIPV victimization scores (victimization-only group), a group characterized by increased EIPV perpetration scores (perpetration-only group), and a group characterized by increased EIPV victimization and perpetration (perpetration-victimization group). Based on prior research (Ulloa & Hammett, 2016), we also hypothesize that membership in the perpetration-victimization group will be associated with the most severe consequences.

Method

Participants and Procedure

Ethical approval for the study was granted by an appropriate institutional ethics committee. The study was pre-registered via Open Science Framework at https://doi.org/10.17605/OSF.IO/UFKP7. The anonymized data that support the findings of this study are available in the SWPS University Repository at https://doi.org/10.58142/swps-ip-ad-01.12.2024. Participants were recruited through the Pollster Research Institute—a certified professional market and public opinion research agency in Poland—using the computer-assisted web interviewing (CAWI) method. The institute maintains a proprietary, actively managed online research panel of registered participants who are continuously profiled on key sociodemographic variables. As part of these standard CAWI quality assurance protocols, the research institute automatically tracked metrics such as survey completion time, response variance in matrix questions, and internal consistency, allowing for the systematic exclusion of low-quality responses prior to the final data delivery (see Figure 1 for specific exclusion criteria and attrition details). Data collection was completed within three weeks.

Figure 1.

Flow diagram representing participants’ attrition from the survey. Note. ^a There were four attention checks (ACs), in which participants were instructed to select a specific answer. Selecting a wrong answer to each question resulted in automatic survey termination. Number of participants who failed each AC: AC1 n = 81, AC2 n = 46, AC3 n = 52, AC4 n = 43. ^b The quality of data was screened by the research company responsible for participant recruitment. Screening included checking the following aspects: time taken to complete the survey (n = 57 cases excluded due to very short completion time), consistency in answering similar or logically connected questions (n = 23 cases excluded due to lack of consistency), and variance in responses to scale elements presented as matrices (n = 62 cases excluded due to lack of variance).

Stratified random sampling procedure was used to recruit a representative sample of ever-partnered women and men aged 18–29 years living in Poland. Sampling was stratified by residential area (urban/rural). All participants had to meet the following criteria: currently residing in Poland, fluent in Polish, history of a relationship that lasted at least 3 months. Details of participant attrition from the study are provided in Figure 1. A total of 3180 individuals completed the survey in full, but 475 were excluded from analyses due to socially desirable responding (the procedure for identifying and excluding socially desirable respondents is described below in the “Measures” subsection). The final sample to be analyzed (N = 2,705) consisted of 1444 (53.4%) women and 1261 (46.6%) men (M age = 24.67 years, SD = 3.35, Mdn = 25). Nearly all participants (n = 2,666, 98.6%) were born in Poland, n = 22 (0.8%) were born in another European country, and n = 17 (0.6%) were born in a non-European country. As for the current place of residence, n = 862 (31.9%) participants came from rural areas, n = 396 (14.6%) declared living in a town of up to 50,000 residents, n = 327 (12.1%) in a town with a population between 50,000 and 150,000 residents, n = 435 (16.1%) in a city with a population between 150,000 and 500,000 residents, and n = 685 (25.3%) participants lived in a city with a population of more than 500,000. For the purpose of further analyses, participants from urban areas were collapsed into a single category. As for the sample distribution by region, n = 674 (24.9%) respondents were from the central region of Poland (voivodeships: Mazovian and Lodz), n = 543 (20.1%) were from the southern region (voivodeships: Silesian and Lesser Poland), n = 430 (15.9%) were from the eastern region (voivodeships: Lublin, Swietokrzyskie, Subcarpathian, and Podlaskie), n = 457 (16.9%) were from the north-western region (voivodeships: West Pomeranian, Lubusz, Greater Poland), n = 254 (9.4%) were from the south-western region (voivodeships: Lower Silesian and Opole), and n = 347 (12.8%) were from the northern region (voivodeships: Pomeranian, Kuyavian-Pomeranian, and Warmian-Masurian). As for the educational attainment, n = 132 (4.9%) participants declared having primary or lower secondary education, n = 190 (7%) completed vocational education, n = 1,235 (45.7%) completed secondary education, and n = 1,148 (42.4%) participants declared having higher education. For the purpose of further analyses, participants without higher education were collapsed into a single category. The majority of the sample (n = 2,063, 76.3%) declared being employed at the time of data collection. Two thousand and sixty-one (n = 2,061, 76.2%) participants declared being in a relationship at the time of data collection.

Measures

Recognition of own EIPV victimization was assessed using a single item: “Have you ever been the victim of emotional intimate partner violence?”. Scoring was dichotomous (0 = no, 1 = yes).

Recognition of own EIPV perpetration was assessed using a single item: “Have you ever been the perpetrator of emotional intimate partner violence?”. Scoring was dichotomous (0 = no, 1 = yes). To avoid priming effects and capture participants’ spontaneous, subjective conceptualization of their experiences, the single-item EIPV recognition questions were intentionally administered prior to the behavioral scales of offline and online EIPV.

Offline EIPV was assessed using the Multidimensional Measure of Emotional Abuse (MMEA; Murphy & Hoover, 1999). The scale was translated into Polish following standard cross-cultural adaptation procedures, utilizing a forward-backward translation method by bilingual experts to ensure semantic equivalence with the original English scales. There are 28 items assessing victimization (MMEA-V) and 28 items assessing perpetration (MMEA-P) grouped under four dimensions (each composed of seven items): RE (refers to efforts to track and monitor one's partner's activities, restrict their freedom, and isolate them from the society), denigration (D; refers to humiliating and belittling one's partner), HW (refers to withholding communication or attention from one's partner to coerce them into behaving in a certain way), and DI (using threats, destruction of property, and other fear-inducing behaviors to achieve submission and compliance). Respondents in the present study were asked to indicate how often in the last 12 months they engaged in specific abusive behaviors against their partner (perpetration) and how often their partner engaged in the same behaviors towards them (victimization): 0 = this has never happened; 1 = once, 2 = twice, 3 = 3–5 times, 4 = 6–10 times, 5 = 11–20 times, 6 = more than 20 times, 7 = never in the past 12 months, but it has happened before. To calculate past-year victimization and perpetration scores, response number 7 was converted to 0 and the items for each scale were summed up. Scores for the total past-year victimization/perpetration scale range between 0–168. To calculate lifetime victimization and perpetration scores, responses number 1–7 were converted to 1 and the items for each scale and subscale were summed up. Scores for the total lifetime victimization/perpetration scale range between 0–28. Higher scores indicate higher levels of offline EIPV victimization/perpetration (MMEA-P: chi square = 1078.41, df = 344, RMSEA = 0.031, 90%CI = 0.029/0.033, CFI = 0.92, TLI = 0.91, SRMR = 0.045, Cronbach's α = .93; MMEA-V: chi square = 1228.64, df = 344, RMSEA = 0.035, 90%CI = 0.033/0.038, CFI = 0.93, TLI = 0.92, SRMR = 0.043, Cronbach's α = .95 [calculated using past-year scores]).

Online EIPV was assessed using the Cyber Dating Abuse Questionnaire (CDAQ; Borrajo et al., 2015). The scale was translated into Polish following standard cross-cultural adaptation procedures, utilizing a forward-backward translation method by bilingual experts to ensure semantic equivalence with the original English scales. There are 20 items assessing victimization (CDAQ-V) and 20 items assessing perpetration (CDAQ-P), grouped into two dimensions: direct aggression (DA; 11 items; refers to using technology-mediated behaviors such as threats and insults with an intention to hurt one's partner) and monitoring/control (MC; nine items; refers to using technology to control one's partner). The response format and methods of total score calculation used in the current study were the same as for the MMEA. Past-year victimization and perpetration scores ranged between 0 and 120 for the total victimization/perpetration scale. Lifetime victimization and perpetration scores ranged between 0 and 20 for the total victimization/perpetration scale. Higher scores indicate higher levels of online EIPV victimization/perpetration (CDAQ-P: chi square = 963.64, df = 169, RMSEA = 0.046, 90%CI = 0.044/0.049, CFI = 0.78, TLI = 0.75, SRMR = 0.084; Cronbach's α = .89; MMEA-V: chi square = 965.15, df = 169, RMSEA = 0.046, 90%CI = 0.044/0.049, CFI = 0.86, TLI = 0.84, SRMR = 0.069; Cronbach's α = .93 [calculated using past-year scores]).

Child abuse and neglect (CAN) was assessed using three items pertaining to emotional abuse (“Did a parent/guardian in your childhood call you names, said mean things to you, or said you were worthless?”), physical abuse (“Did a parent/guardian in your childhood physically hurt you in any way [e.g., hit, beat, kick]?”), and neglect (“In your childhood, did you have to look after yourself because a parent/guardian drank too much alcohol, took drugs, or was completely uninterested in you?”) (Debowska & Boduszek, 2017). Scoring on individual items was dichotomous (1 = no, 2 = yes). Scores were summed across the three items (possible range of scores: 3–6). Higher scores indicate increased levels of childhood violence.

Current mental distress was assessed using the four-item Patient Health Questionnaire for Depression and Anxiety (PHQ-4; Kroenke et al., 2009). Participants were asked to indicate how often (from 0 = not at all to 3 = nearly every day) over the last two weeks they had been bothered by symptoms of depression (e.g., “Little interest or pleasure in doing things”) and anxiety (e.g., “Not being able to stop or control worrying”). Scores were summed across the four items (possible range of scores: 0–12). Increased scores indicate higher current mental distress (Cronbach's α = .86).

Self-esteem was measured using the Single-Item Self-Esteem Scale (Robins et al., 2001). The item was designed to assess global self-esteem. Respondents were asked to react to the single item (“I have high self-esteem”) using a 7-point Likert scale, ranging from 1 (not very true of me) to 7 (very true of me). Higher scores indicate increased self-esteem levels.

Emotional and social loneliness were assessing using the 6-item DeJong Gierveld Loneliness Scale (De Jong Gierveld & Van Tilburg, 2006; Grygiel et al., 2013), which assesses the feeling of missing an intimate relationship (emotional loneliness; e.g.,: “I experience a general sense of emptiness”) and missing a wider social network (social loneliness; e.g.,: “There are plenty of people I can rely on when I have problems”). Responses are scored on a five-point Likert scale (1 = yes!, 2 = yes, 3 = more or less, 4 = no, 5 = no!). Scores were summed across the three emotional loneliness items and social loneliness items (possible range of scores for each subscale: 3–12). Increased scores indicate higher loneliness (Cronbach's α for emotional loneliness = 0.80 and social loneliness = 0.83).

Global health status (Hays et al., 2015) was assessed with a single question: “In general, how would you rate your physical health?” with five response options: 1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent.

Demographics questionnaire included questions about age (reported in years), level of education, employment status, sexual orientation, country of birth, place of residence, and history of mental health diagnosis (present/absent).

Social desirability. To control for social desirability bias, we used three items from the Eysenck Personality Questionnaire-Revised Lie scale (Eysenck et al., 1985). These were: (1) “Are all your habits good and desirable ones?”; (2) “Have you ever taken anything (even a pin or button) that belonged to someone else?”; (3) “Have you ever said anything bad or nasty about anyone?”. The items were scored using a Yes/No format. Total scores ranged from 0 to 3, with higher scores indicating greater social desirability in responses. Socially desirable responding was operationalized as an overall score at or above 2 (≥ 2) (as previously used by Debowska et al., 2022). In the current study, 475 participants were classified as socially desirable respondents and were excluded from analyses.

Data Analytic Plan

Descriptive statistics were calculated using IBM SPSS Statistics, Version 29. To examine associations between offline and online EIPV victimization and perpetration scores as well as associations between offline and online EIPV victimization/perpetration and recognition of own EIPV victimization/perpetration, we performed Spearman's rank-order correlations (past-year scores were used for all EIPV variables). To establish profiles of EIPV victimization and perpetration and associated risk factors and consequences, we performed latent profile analysis (LPA) with covariates using Mplus version 7.4 (Muthén & Muthén, 1998–2015). Goodness-of-fit statistics used for deciding on the optimal number of profiles included: the Akaike Information Criterion (AIC; Akaike, 1974), Bayesian Information Criterion (BIC; Schwarz, 1978), sample size adjusted BIC (_SSABIC; Sclove, 1987), Lo–Mendell–Rubin's adjusted likelihood ratio test (LMR; Lo et al., 2001), and entropy (Ramaswamy et al., 1993). AIC, BIC and _SSABIC are goodness of fit indices used to compare competing models. Lower values indicate better-fitting models and extraction of latent profiles should cease when these indices reach their lowest values. The LMR was used to assess the improvement in fit between alternative models. A non-significant value (p > .05) suggests that the model with one fewer group provides a more parsimonious fit to the data. Entropy values show the ability of the model to correctly classify participants. Possible entropy values range from 0 to 1, with higher values indicating better classification. It has been recommended that researchers implement a cut-off value of 0.80, indicating that individuals are allocated to the correct group 80% of the time (Clark & Muthén, 2009). Prior to running all analyses, we checked the required assumptions (e.g., sample size, missing values, outliers, normality, linearity, and multicollinearity). Unmet assumptions were dealt with appropriately.

Results

Descriptive Statistics for EIPV Variables

Table 1 contains descriptive information on all EIPV variables for the total sample as well as females and males separately. In addition, the table provides information on sex differences in the scores on EIPV variables. There were no statistically significant differences between the sexes on offline victimization. However, women, compared with men, were found to perpetrate more past-year offline and lifetime online EIPV. Men scored statically significantly higher than women on past-year and lifetime online victimization.

Table 1.

Descriptive Statistics for All Emotional Intimate Partner (EIPV) Variables in the Total Sample (N = 2,705), Among Women (n = 1,444), and Among Men (n = 1,261) As Well As Sex Differences.

Type of Abuse	Total Sample M (SD), Mdn, Min-Max	Women M (SD), Mdn, Min-Max	Men M (SD), Mdn, Min-Max
Past-year EIPV victimization offline	25.21 (27.18), 17, 0–158	24.60 (27.12), 16, 0–158	25.91 (27.25), 18, 0–154
Lifetime EIPV victimization offline	10.11 (7.12), 9, 0–28	9.91 (6.89), 9, 0–28	10.34 (7.36), 9, 0–28
Past-year EIPV perpetration offline *	19.33 (20.66), 14, 0–148	19.57 (19.64), 15, 0–133	19.06 (21.79), 13, 0–148
Lifetime EIPV perpetration offline	8.77 (6.46), 8, 0–28	8.72 (5.84), 8, 0–28	8.83 (7.10), 8, 0–28
Past-year EIPV victimization online **	7.95 (15.02), 2, 0–118	6.66 (13.59), 1, 0–118	9.43 (16.39), 3, 0–110
Lifetime EIPV victimization online **	3.28 (4.49), 2, 0–20	2.86 (4.00), 1, 0–20	3.75 (4.94), 2, 0–20
Past-year EIPV perpetration online	6.89 (11.71), 3, 0–105	6.59 (10.39), 3, 0–92	7.22 (13.05), 2, 0–105
Lifetime EIPV perpetration online *	3.02 (3.89), 2, 0–20	2.91 (3.34), 2, 0–20	3.15 (4.44), 2, 0–20

Note. * indicates a statistically significant difference between women and men, where women scored higher than men (comparisons were made using Mann–Whitney U tests to account for non-normal distribution of scores); ** indicates a statistically significant difference between women and men, where men scored higher than women (comparisons were made using Mann–Whitney U tests to account for non-normal distribution of scores).

Descriptive Statistics for Non-EIPV Variables

Descriptive statistics for the full sample and separately for female and male participants are presented in Table 2. The data show that approximately 1 in 3 women and 1 in 4 men acknowledge themselves as EIPV victims. Fewer participants recognized themselves as EIPV perpetrators: 1 in 10 women and 1 in 14 men. Our sample also offers a good representation of non-heterosexual individuals (13.4% women and 15.5% men), who are frequently neglected or even actively excluded from IPV research. Also noteworthy, 16.7% of the community-based participants (18.6% women and 14.6%) reported having a psychiatric diagnosis.

Table 2.

Descriptive Statistics for Categorical, Ordinal, and Non-EIPV Continuous Variables Presented for the Full Sample, Female Participants, and Male Participants.

Sample Characteristics		Full Sample (N = 2,705)	Female Participants (n = 1,444)	Male Participants (n = 1,261)
Categorical variables
Recognition of own EIPV victimization
	Yes	28.5% (772)	33.8% (488)	22.5% (284)
Recognition of own EIPV perpetration
	Yes	8.9% (240)	10.5% (152)	7% (88)
Education level
	Higher	42.4% (1148)	52.2% (754)	31.2% (394)
	Non-higher	57.6% (1557)	47.8% (690)	68.8% (867)
Place of residence
	Urban	68.1% (1843)	64.1% (926)	72.7% (917)
	Rural	31.9% (862)	35.9% (518)	27.3% (344)
Psychiatric diagnosis
	Present	16.7% (453)	18.6% (269)	14.6% (184)
Sexual orientation
	Heterosexual	85.6% (2315)	86.6% (1250)	84.5% (1065)
	Non-heterosexual	14.4% (390)	13.4% (194)	15.5% (196)
Ordinal variables
Global health status
	Median	3	3	3
	Mode	3	3	4
	Min-Max	1–5	1–5	1–5
Self-esteem
	Median	4	3	4
	Mode	3	3	4
	Min-Max	1–7	1–7	1–7
Continuous variables
Age
	M (SD)	24.67 (3.35)	24.92 (3.27)	24.38 (3.42)
	Median	25	26	25
	Min-Max	18–29	18–29	18–29
Child abuse and neglect
	M (SD)	4.06 (1.08)	4.11 (1.07)	4.00 (1.09)
	Median	4	4	4
	Min-Max	3–6	3–6	3–6
Current mental distress
	M (SD)	4.41 (3.21)	4.83 (3.25)	3.93 (3.10)
	Median	4	4	3
	Min-Max	0–12	0–12	0–12
Emotional loneliness
	M (SD)	9 (2.81)	9.22 (2.82)	8.75 (2.80)
	Median	9	9	9
	Min-Max	3–15	3–15	3–15
Social loneliness
	M (SD)	9.05 (2.91)	9.19 (2.99)	8.90 (2.81)
	Median	9	9	9
	Min-Max	3–15	3–15	3–15

Note. EIPV = emotional intimate partner violence.

Prevalence of EIPV Victimization and Perpetration

We examined the prevalence of each type of EIPV (past-year victimization offline, past-year victimization online, past-year perpetration offline, past-year perpetration online, lifetime victimization offline, lifetime victimization online, lifetime perpetration offline, lifetime perpetration online) through transforming MMEA and CDAQ total score variables into three-level categorical variables to indicate three levels of incidence of any EIPV abuse: never, once or twice, more than twice. In Table 3 we present lifetime and past-year prevalence of EIPV victimization and perpetration offline and online among the total sample and separately for women and men. These statistics reveal that the majority of men and women (around 80%) were victimized and perpetrated EIPV offline more than twice. The statistics for past-year and lifetime prevalence are similar. EIPV victimization and perpetration online is a rarer yet still frequent occurrence, affecting around 40% of men and women more than twice in the lifetime and in the past year. Fewer participants reported having experienced EIPV once or twice. The discrepancy between EIPV happening once or twice compared with more than twice was particularly pronounced for offline abuse. This means that once an EIPV behavior occurs, it is either likely to be repeated or co-occur with other EIPV behaviors. No prominent differences in how women and men experience EIPV were noted, indicating that EIPV affects the two sexes at comparable rates. Under study objective 1, we hypothesized that at least one in two men and women would report lifetime EIPV experiences and that the prevalence rates would be similar across the sexes and these predictions are supported by the current data.

Table 3.

Past-Year and Lifetime Prevalence of Emotional Intimate Partner Violence (EIPV) Offline and Online in the Total Sample (N = 2,705) and Among Women (n = 1,444) and Men (n = 1,261) Separately.

Type of Abuse		Total Sample	Women	Men
Past-year EIPV victimization offline
	No experience	12.6% (n = 340)	11.9% (n = 172)	13.3% (n = 168)
	Once or twice	4.4% (n = 120)	4.7% (n = 69)	4% (n = 51)
	More than twice	83% (n = 2,245)	83.3% (n = 1,203)	82.6% (n = 1,042)
Past-year EIPV victimization online
	No experience	44.5% (n = 1,204)	47.6% (n = 688)	40.9% (n = 516)
	Once or twice	10% (n = 271)	11.3% (n = 163)	8.6% (n = 108)
	More than twice	45.5% (n = 1,230)	41.1% (n = 593)	50.5% (n = 637)
Past-year EIPV perpetration offline
	No experience	12.6% (n = 342)	10.2% (n = 148)	15.4% (n = 194)
	Once or twice	5.7% (n = 154)	5.4% (n = 78)	6% (n = 76)
	More than twice	81.7% (n = 2,209)	84.3% (n = 1,218)	78.6% (n = 991)
Past-year EIPV perpetration online
	No experience	39.1% (n = 1,059)	37.5% (n = 542)	41% (n = 517)
	Once or twice	10.8% (n = 291)	10.2% (n = 148)	11.3% (n = 143)
	More than twice	50.1% (n = 1,355)	52.2% (n = 754)	47.7% (n = 601)
Lifetime EIPV victimization offline
	No experience	6.4% (n = 172)	6.3% (n = 91)	6.4% (n = 81)
	Once or twice	7.3% (n = 198)	6.9% (n = 99)	7.9% (n = 99)
	More than twice	86.3% (n = 2,335)	86.8% (n = 1,254)	85.7% (n = 1,081)
Lifetime EIPV victimization online
	No experience	33.2% (n = 899)	35.2% (n = 509)	30.9% (n = 390)
	Once or twice	27.2% (n = 735)	28.8% (n = 416)	25.3% (n = 319)
	More than twice	39.6% (n = 1,071)	35.9% (n = 519)	43.8% (n = 552)
Lifetime EIPV perpetration offline
	No experience	6.6% (n = 178)	4.6% (n = 67)	8.8% (n = 111)
	Once or twice	9.3% (n = 252)	8.8% (n = 127)	9.9% (n = 125)
	More than twice	84.1% (n = 2,275)	86.6% (n = 1,250)	81.3% (n = 1,025)
Lifetime EIPV perpetration online
	No experience	29.7% (n = 804)	27.2% (n = 393)	41% (n = 517)
	Once or twice	28.3% (n = 765)	28.9% (n = 417)	11.3% (n = 143)
	More than twice	42% (n = 1,136)	43.9% (n = 634)	47.7% (n = 601)

Note. EIPV = Emotional intimate partner violence.

Correlations Between Offline and Online EIPV Victimization/Perpetration

Correlations between offline and online EIPV victimization were statistically significant (p < .001) and moderate (interpreted as values between 0.38 and 0.67; see Schober et al., 2018) for the full sample (rho = 0.63) and for women (rho = 0.60) and men (rho = 0.67) separately. Similarly, associations between offline and online EIPV perpetration scores were statistically significant and moderate for the full sample (rho = 0.59) and for women (rho = 0.58) and men (rho = 0.60) separately. These results provide support for our predictions listed under study objective 2, referring to correlations between offline and online EIPV.

Correlations Between EIPV Victimization and Perpetration

Offline EIPV victimization scores formed strong (interpreted as values between 0.68 and 0.88; see Schober et al., 2018) to very strong (interpreted as ≥ 0.89), statistically significant correlations (p < .001) with offline EIPV perpetration scores in the total sample (rho = 0.82), among females (rho = 0.79), and among males (rho = 0.86). Correlations between offline EIPV victimization and online EIPV perpetration were statistically significant (p < .001) and moderate for the total sample (rho = 0.55), females (rho = 0.53), and males (rho = 0.57).

Online EIPV victimization was statistically significantly (p < .001) and strongly associated with online EIPV perpetration for the total sample (rho = 0.75), females (rho = 0.72), and males (rho = 0.80). Online EIPV victimization and offline EIPV perpetration were significantly (p < .001) associated with each other to a moderate degree for the total sample (rho = 0.55), females (rho = 0.51), and males (rho = 0.61). Taken together, these findings indicate that our predictions listed under objective 3 have been met.

Recognition of Own EIPV Victimization and Perpetration

First, offline and online EIPV victimization scores were correlated with a single item own EIPV victimization recognition. Results indicate that both offline and online EIPV victimization was statistically significantly (p < .001) yet weakly (interpreted as values between 0.10 and 0.37; see Schober et al., 2018) associated with own EIPV victimization recognition in the total sample (offline EIPV: rho = 0.22; online EIPV: rho = 0.13), female sample (offline EIPV: rho = 0.25; online EIPV: rho = 0.15), and the male sample (offline EIPV: rho = 0.19; online EIPV: rho = 0.14).

Next, offline and online EIPV perpetration total scores were correlated with a single item own EIPV perpetration recognition. Findings demonstrate that associations between both offline and online EIPV victimization and own EIPV perpetration recognition were statistically significant (p < .001) yet weak among all participants (offline EIPV: rho = 0.18; online EIPV: rho = 0.13), women (offline EIPV: rho = 0.20; online EIPV: rho = 0.16), and men (offline EIPV: rho = 0.16; online EIPV: rho = 0.10). All in all, these findings indicate that our hypotheses listed under objective 4 have been met.

Profiles of Offline and Online EIPV Victimization and Perpetration

The fit statistics for the LPA of EIPV are presented in Table 4. The lowest AIC, BIC, and _SSABIC values were recorded for the 4-group solution. However, the LMR value for the 4-group model was statistically non-significant, indicating that the model with one fewer group (i.e., the 3-group model) provides a more parsimonious fit to the data. Entropy value was also higher for 3-group model compared with the 4-group model. Finally, the additional profile in the 4-group model was extremely small in terms of group membership (1% of participants) and hence was discarded as a statistical anomaly (Hipp & Bauer, 2006).

Table 4.

Fit Indices for the Latent Profile Analysis of Offline and Online Emotional Intimate Partner Violence Victimization and Perpetration (N = 2,705).

Model	AIC	BIC	_SSABIC	LMR	p	Entropy
2 classes	877374.33	87521.91	87422.47	5557.15	<.001	.97
3 classes	85146.49	85394.42	8526.97	2245.13	.05	.97
4 classes	83618.54	83966.81	83779.35	1550.42	.26	.95

Note. AIC = Akaike information criterion; BIC = Bayesian information criterion; BLRT = Bootstrapped Likelihood Ratio Test; _SSABIC = sample size adjusted BIC; LMR = Lo–Mendell–Rubin's adjusted likelihood ratio test. Boldface type suggests the best-fitting model.

Figure 2 shows the profile plot for the 3-group solution. Group 1 (82.9% of participants) is the largest group. It is characterized by low mean scores on all EIPV variables and is labeled the “low EIPV” group. Group 2 (14.1% of participants) is characterized by moderate mean scores on EIPV variables and is labeled the “moderate EIPV” group. Finally, group 3 (3% of participants) is characterized by high mean scores on all EIPV variables and is labeled the “high EIPV” group. Table 5 displays estimated means, standard errors, and 95% confidence intervals for EIPV indicators by latent group.

Figure 2.

Latent profile analysis plot of offline and online emotional intimate partner violence (EIPV) victimization and perpetration (N = 2,705). Note. Dashed line—Group 1 = “low EIPV” group (82.9% of cases); Dotted line—Group 2 = “moderate EIPV” group (14.1%); Solid line—Group 3 = “high EIPV” group (3%). EIPV-P = Emotional intimate partner violence perpetration; EIPV-V = Emotional intimate partner violence victimization.

Table 5.

Estimated Means, Standard Errors, and 95% Confidence Intervals for EIPV Indicators by Latent Group.

Indicator	Group 1: Low EIPV (82.9%)	Group 2: Moderate EIPV (14.1%)	Group 3: High EIPV (3.0%)
Offline EIPV-V	M = 13.61, SE = 0.44, 95%CI (12.75, 14.47)	M = 43.91, SE = 2.22, 95%CI (39.56, 48.26)	M = 94.55, SE = 5.65, 95%CI (83.48, 105.62)
Offline EIPV-P	M = 16.65, SE = 0.60, 95%CI (15.47, 17.83)	M = 66.71, SE = 3.54, 95%CI (59.77, 73.65)	M = 100.91, SE = 3.65, 95%CI (93.76, 108.06)
Online EIPV-V	M = 3.56, SE = 0.19, 95%CI (3.19, 3.93)	M = 19.36, SE = 1.22, 95%CI (16.97, 21.75)	M = 64.56, SE = 7.10, 95%CI (50.64, 78.48)
Online EIPV-P	M = 3.13, SE = 0.21, 95%CI (2.72, 3.54)	M = 28.59, SE = 2.42, 95%CI (23.85, 33.33)	M = 71.88, SE = 4.60, 95%CI (62.86, 80.90)

Note. EIPV = Emotional intimate partner violence; EIPV-P = Emotional intimate partner violence perpetration; EIPV-V = Emotional intimate partner violence victimization.

The associations between EIPV group membership and external criteria were calculated using multinominal logistic regression analyses (see Table 6). Results indicate that members of the “moderate EIPV” group, compared with members of the “low EIPV group,” were significantly more likely to be less educated, experience higher current mental distress, and report more emotional loneliness. Members of the “high EIPV” group, compared with members of the “low EIPV” and “medium EIPV” groups, were significantly more likely to be male.

Table 6.

Results From Multinomial Logistic Regression (Associations Between the 3 Latent Groups and External Criteria).

Variables	2 Vs 1 OR (95% CI)	3 Vs 1 OR (95% CI)	3 Vs 2 OR (95% CI)
Sex	1.14 (0.98–6.27)	2.97** (2.38–5.71)	2.61** (2.05- 4.32)
Age	0.97 (0.87–1.08)	0.94 (0.75–1.17)	0.97 (0.76–1.23)
Level of education	0.68** (0.61–0.76)	0.93 (0.75–1.16)	1.40 (0.92–1.78)
Place of residence	0.99 (0.89–3.10)	1.21 (0.97–3.51)	1.21 (0.95–2.54)
Sexual orientation	1.08 (0.97–2.20)	1.22 (0.98–2.52)	1.13 (0.89–2.44)
History of CAN	1.07 (0.96–2.19)	1.10 (0.88–2.37)	1.02 (0.80–1.30)
Psychiatric diagnosis	1.04 (0.93–2.16)	1.14 (0.91–3.42)	1.09 (0.86–2.39)
Global health status	0.93 (0.83–1.04)	0.82 (0.66–1.02)	0.87 (0.68–1.11)
Current mental distress	1.13*** (1.01–1.26)	1.14 (0.91–1.42)	1.01 (0.79–1.28)
Self-esteem	1.09 (0.98–1.22)	1.15 (0.92–1.44)	1.05 (0.83–1.33)
Emotional loneliness	1.13*** (1.01–1.26)	1.06 (0.85–1.32)	0.94 (0.74–1.19)
Social loneliness	1.02 (0.92–1.14)	1.02 (0.82–1.27)	1.01 (0.79–1.28)

Note. Group 1 = “low EIPV” group (reference group); Group 2 = “moderate EIPV” group; Group 3 = “high EIPV” group; OR = Odds Ratio; 95% CI = Confidence Interval. CAN = Child abuse and neglect. EIPV = Emotional intimate partner violence.

p < .01.

***

p < .001.

Discussion

The overall aim of the present study was to provide a comprehensive assessment of offline and online EIPV victimization and perpetration among young women and men (aged 18–29 years) in Poland. To the best of our knowledge, based on our review of the existing literature, the specific topic of EIPV in the Polish population has remained largely unexplored. Therefore, the current research represents the first comprehensive study of EIPV utilizing a large, representative sample of Polish young adults.

In line with objective 1, we established the past-year and lifetime prevalence rates of offline and online EIPV victimization and perpetration among the total sample and among women and men separately. Based on prior research (Doroszewicz & Forbes, 2008; Spencer et al., 2024), we predicted that at least 1 in 2 men and women in the current sample would report lifetime EIPV victimization and perpetration experiences. The present results exceed our initial expectations and largely mirror the rates provided in Doroszewicz and Forbes’ (2008) study, where 77% of men and 89% of women were found to experience EIPV in a dating relationship. Specifically, we found that in excess of 80% of both men and women in our sample were victimized and perpetrated offline EIPV more than twice in their lifetime. The rates of past-year prevalence were only slightly lower. Notably, considering lifetime prevalence, only 1 in 16 women and men reported having experienced no offline EIPV victimization and 1 in 22 women and 1 in 11 men reported no EIPV perpetration history. These findings indicate that EIPV offline is a frequent event and that EIPV may be reciprocal in the relationships of young adults in Poland. EIPV victimization and perpetration online was a rarer yet still frequent occurrence, affecting around 40% of men and women more than twice in the lifetime and in the past year. The rates of online EIPV reported here are reflective of cyber partner abuse rates previously reported for American college students (Wolford-Clevenger et al., 2016). It is important to contextualize the high prevalence estimates observed in the current study and prior similar studies. The instruments utilized in this research employ an act-based approach, measuring the occurrence of specific psychological and digital behaviors (e.g., hostile withdrawal, denigration, digital monitoring) rather than the resulting trauma, fear, or systematic coercive control. While these behaviors undeniably constitute emotional abuse within psychological and relational frameworks, they also represent highly prevalent, often normalized, maladaptive conflict resolution tactics among young adults. Consequently, our estimates naturally exceed the prevalence rates typically associated with public health or legal frameworks of intimate partner violence, which generally rely on much higher thresholds of severity, demonstrable psychological injury, or chronic systemic control to define abuse.

In addition, gender comparisons across both offline and online EIPV modalities demonstrate nuanced differences in how men and women navigate and experience psychological aggression. Based on past-year and lifetime prevalence rates, men were more likely than women to be emotionally victimized online. Women reported higher levels of offline and online EIPV perpetration than men. These results are in line with prior research indicating that when women commit IPV, it is predominantly in the form of psychological abuse (Williams et al., 2008). These findings may indicate that IPV, similarly to other forms of aggression and violence (Archer, 2004), conforms to traditional gender norms, whereby it is less socially acceptable for women to use overtly aggressive behaviors (such as physical violence) and so they tend to revert to the more hidden aggression forms (such as monitoring one's partner's activities and withholding communication to coerce one's partner into behaving in a certain way). Although some previous research rejected the assumptions of social role theory in the context of IPV behaviors (e.g., Lascorz et al., 2018), Polish society is characterized by traditional values and a relatively strong gender stereotype (Mandal et al., 2012), which can affect the behavior of young adults in intimate relationships. Overall, EIPV prevalence among both women and men reported here are alarmingly high and require immediate attention from appropriate government agencies and non-governmental organizations. Nevertheless, while our findings indicate high frequencies of EIPV perpetration among women and notable rates of online victimization among men, it is imperative to interpret these gender dynamics with caution to avoid overgeneralization. Consistent with broader IPV literature, the frequency of act-based behaviors does not inherently equate to the severity of harm, fear, or trauma experienced by the victim. Research consistently highlights that even when women and men engage in similar rates of psychological aggression, the contextual factors, the severity of abuse victimization, underlying motivations (such as defensive behavior versus coercive control), and resulting psychological impacts often manifest asymmetrically (Ansara & Hindin, 2010; Carbone-López et al., 2006; Johnson, 2006; Lagdon et al., 2023).

To address our second and third objectives, subsequent analyses investigated the correlations between EIPV victimization and perpetration, along with the crossover between offline and online EIPV. Based on prior research, we anticipated moderate to strong associations between offline and online EIPV (e.g., Gilbar et al., 2023; Schokkenbroek et al., 2022) and between EIPV victimization and perpetration (e.g., Jaureguizar et al., 2024; Schokkenbroek et al., 2022). In line with these predictions, correlations between offline and online EIPV victimization as well as between offline and online EIPV perpetration were statistically significant and moderate for the full sample and for women and men separately. This demonstrates a significant convergence of abusive tactics, wherein EIPV behaviors concurrently manifest in both in-person and digital settings. Analyses of EIPV reciprocity revealed strong within-modality correlations (offline victimization with offline perpetration, and online victimization with online perpetration) alongside moderate cross-modality associations. These findings substantiate the perpetrator–victim phenomenon, illustrating that individuals in our sample frequently operate as both targets and perpetrators of abuse. One possible explanation of the phenomenon is that young people do not recognize EIPV behaviors as abusive. Indeed, our findings highlight a profound disconnect between the objective experience of EIPV and the subjective recognition of abuse. While over 80% of our sample engaged in or experienced EIPV in the past year, only a fraction self-identified as victims (1 in 3 women; 1 in 4 men) or perpetrators (1 in 10 women; 1 in 14 men). This weak association, which aligns with our predictions (see objective 4), can likely be attributed to the normalization and romanticization of toxic relational behaviors. As past research suggests, many psychological control tactics—such as intense monitoring or jealousy—are frequently culturally coded or disguised as desirable expressions of deep affection and protectiveness (Minto et al., 2022). Consequently, young adults often lack the cognitive schemas necessary to accurately label these normalized dynamics as intimate partner violence.

Finally, in line with objective 5, we established three meaningful profiles of offline and online EIPV victimization/perpetration using latent profile analysis (LPA). Based on the mean scores on all EIPV variables, the groups were labeled: “low EIPV” (82.9% of participants), “moderate EIPV” (14.1% of participants), and “high EIPV” groups (3% of participants). The highest mean scores for all groups were recorded on offline EIPV victimization, followed by offline EIPV perpetration, online EIPV victimization, and online EIPV perpetration. The results indicate that the groups differ quantitatively but not qualitatively from one another, which only partly supports our predictions. Namely, the solution did not include any victimization-only or perpetration-only groups, which were expected to be uncovered. This lends support to some prior research demonstrating that EIPV is likely to be bidirectional (O’Leary & Slep, 2012; Wagers et al., 2021). Consistent with our findings, Grest et al. (2018) found that among Latino adolescents, the only distinct EIPV profile that emerged was characterized by both victimization and perpetration. The strong co-occurrence of EIPV victimization and perpetration can likely be understood through a dyadic or systemic lens. In many young adult relationships, psychological aggression may become a normalized, bidirectional pattern of conflict resolution, where one partner's toxic behaviors (e.g., hostile withdrawal or denigration) trigger a retaliatory or defensive abusive response from the other, creating a cycle of mutual emotional abuse (Dokkedahl & Elklit, 2019). Further, our “high EIPV” group, compared with the remaining groups, was significantly more likely to include male participants. Mean scores for this group were particularly high for victimization variables. Members of the “moderate EIPV” group, compared with the “low EIPV” group members, were less educated and experienced more current mental distress and emotional loneliness, which highlights some important risk factors as well as consequences of EIPV experience. Prevention developers can leverage these specific demographic and psychosocial markers to improve screening protocols and allocate resources toward the most vulnerable subgroups. The lack of clear distinguishing features of the “high EIPV” group could be explained by the extremely low group membership (3%). Future studies should aim to recruit participants with heightened EIPV victimization and perpetration to build a better understanding of risk factors and consequences associated with such experiences.

The identification of these three distinct profiles—low, moderate, and high EIPV—significantly enhances our conceptual and practical understanding of emotional partner abuse. Crucially, the absence of “victim-only” or “perpetrator-only” groups, alongside the lack of modality-specific profiles (e.g., “online-only” abuse), suggests that EIPV among young adults operates primarily on a continuum of severity rather than existing as distinct behavioral typologies. The emergence of the “moderate” and “high” EIPV groups, representing approximately 17% of our sample, highlights a substantial subgroup trapped in intense, bidirectional cycles of psychological aggression that span both physical and digital environments. From an applied perspective, these profiles underscore that clinical interventions and prevention programs for young adults must move beyond binary victim/perpetrator paradigms. Instead, therapeutic efforts should be systemically oriented, focusing on relational dynamics, teaching mutual de-escalation, establishing healthy digital boundaries, and promoting constructive conflict resolution to disrupt these deeply entrenched cycles of mutual abuse. In addition, prior IPV prevention tended to focus on men as perpetrators. Our study shows that women and men are likely to perpetrate EIPV and so both sexes should be the targets of EIPV prevention efforts.

Limitations and Future Directions

This study has several limitations that should be noted. First, we relied on self-reports which are subject to social desirability bias in responding. Indeed, prior research demonstrated that IPV scholarship is affected by a specific manifestation of the social desirability bias—impression management, which refers to the deliberate presentation of false information in order to create more favorable impressions. Reports of both IPV victimization and perpetration were found to be prone to social desirability bias, with a greater effect found for IPV perpetration reports (Visschers et al., 2017). However, to alleviate this issue, we assessed socially desirable responding and excluded participants who expressed the tendency to present themselves in a favorable light. Second, we were aiming to recruit 40% of all participants from rural areas, but this proved difficult, especially in the case of men. Our final sample consisted of 35% of women and 27% of men who lived in rural areas of Poland. Future research should explore ways to reach this demographic, especially since rural residence is a known risk factor for IPV. Nevertheless, we managed to recruit participants from all regions and so EIPV experiences reported here are representative for young adults in the whole country. Future research would also benefit from qualitative or mixed-methods approaches to explore whether and when young adults explicitly label their current or former relationships as “abusive” following EIPV exposure, as subjective labeling may differ from the recognition of specific harmful behaviors. Furthermore, while our data-driven latent profile analysis primarily uncovered groups characterized by the co-occurrence of victimization and perpetration (the perpetrator–victim phenomenon), future research should investigate whether distinct “victim-only” and “perpetrator-only” profiles might emerge in different contexts, such as clinical or forensic samples.

Footnotes

ORCID iDs

Agata Debowska

Daniel Boduszek

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Science Centre Poland (grant number 2023/07/X/HS6/00881).

Declaration of Conflicting Interests

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

Author Biographies

Agata Debowska, PhD, is a Chartered Psychologist (CPsychol) with the British Psychological Society and a Fellow of the Higher Education Academy (HEA) in the UK. She is an Associate Professor at SWPS University (Warsaw, Poland) and serves as the Director of SWPS University Center for Open Education. Her research focuses on intimate partner violence (IPV), child abuse and neglect (CAN), violence prevention, and mental health.

Urszula Bojnowska, B.A., is a postgraduate psychology student at SWPS University (Poland). Her academic interests focus on intimate partner violence, particularly emotional intimate partner violence and its psychological correlates. She plans to pursue doctoral studies in psychology, followed by clinical training in cognitive-behavioral therapy (CBT). Her broader interests include the prevention of interpersonal violence and the development of evidence-based psychological interventions.

Michał Sienkiewicz, M.A., holds a Master's degree in psychology. He has professional experience working on a domestic violence support helpline and a crisis helpline for adults experiencing psychological distress. He has provided individual consultations for people affected by domestic violence, co-facilitated a support group for adolescent girls, and conducted individual parental consultations. His professional interests include crisis intervention, psychological support for individuals affected by violence, and family-focused interventions.

Olga Krawczyk, M.A., is a counseling psychologist, International Baccalaureate (IB) diploma teacher, and a psychotherapist in training. She works primarily with adolescents and young adults. Her main research interests involve cognitive and clinical psychology, including gender differences in narcissistic personality, as well as intimate partner violence and issues concerning public health.

Aleksandra Chłopecka, B.A., is a postgraduate psychology student at SWPS University of (Warsaw, Poland). She has experience working in educational settings with children and adolescents and is particularly interested in pursuing clinical practice in the future. Her focus is on working with adolescents and young adults experiencing a range of mental health crises, including eating disorders, anxiety disorders, and obsessive-compulsive disorder (OCD). Her academic interests also include trauma-focused interventions, intimate partner violence, and violence prevention.

Daniel Boduszek, Ph.D, is a Certified Clinical Psychologist, Professor of Criminal Psychology at SWPS University, and a Visiting Professor of Forensic Psychology at Loughborough University. His main scientific interests include violence against women and children, psychopathy, criminal cognitions & criminal social identity, suicide and mental health, as well as the issues of homicide, recidivism, and prisonization.

References

Akaike

(1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19(6), 716–723. https://doi.org/10.1109/TAC.1974.1100705

Ansara

D. L.

Hindin

M. J.

(2010). Exploring gender differences in the patterns of intimate partner violence in Canada: A latent class approach. Journal of Epidemiology & Community Health, 64(10), 849–854. https://doi.org/10.1136/jech.2009.095208

Archer

(2004). Sex differences in aggression in real-world settings: A meta-analytic review. Review of General Psychology, 8(4), 291–322. https://doi.org/10.1037/1089-2680.8.4.291

Barnes

Szilassy

Herbert

Heron

Feder

Fraser

, … Barter

(2022). Being silenced, loneliness and being heard: Understanding pathways to intimate partner violence & abuse in young adults. a mixed-methods study. BMC Public Health, 22(1), 1562. https://doi.org/10.1186/s12889-022-13990-4

Borrajo

Gámez-Guadix

Pereda

Calvete

(2015). The development and validation of the cyber dating abuse questionnaire among young couples. Computers in Human Behavior, 48, 358–365. https://doi.org/10.1016/j.chb.2015.01.063

Breiding

M. J.

(2015). Prevalence and characteristics of sexual violence, stalking, and intimate partner violence victimization—National intimate partner and sexual violence survey, United States, 2011. American Journal of Public Health, 105(4), e11–e12. https://doi.org/10.2105/AJPH.2015.302634

Brunell

(2005). Marginality and the new geography of domestic violence policy in post- communist Poland. Gender, Place & Culture, 12(3), 293–316. https://doi.org/10.1080/09663690500202533

Carbone-López

Kruttschnitt

Macmillan

(2006). Patterns of intimate partner violence and their associations with physical health, psychological distress, and substance use. Public Health Reports, 121(4), 382–392. https://doi.org/10.1177/003335490612100406

Clark

S. L.

Muthén

(2009). Relating latent class analysis results to variables not included in the analysis. University of California. https://www.statmodel.com/download/relatinglca.pdf .

10.

Debowska

Boduszek

(2017). Child abuse and neglect profiles and their psychosocial consequences in a large sample of incarcerated males. Child Abuse & Neglect, 65, 266–277. https://doi.org/10.1016/j.chiabu.2016.12.003

11.

Debowska

Horeczy

Boduszek

Dolinski

von Bastian

C. C.

(2022). Development and validation of a stress response measure: The Daily Stress Response Scale (DSRS). Health Psychology Report, 10(3), 238–248. https://doi.org/10.5114/hpr.2022.116812

12.

Debowska

Inglot

Piasek

Sokol

Horeczy

Hales

G. K.

Boduszek

(2025). Testing the spillover effect of intimate partner violence victimization on emotionally abusive and harsh parenting practices: The application of propensity score matching. Journal of Interpersonal Violence, 40(5-6), 1199–1220. https://doi.org/10.1177/08862605241258998

13.

De Jong-Gierveld

van Tilburg

T. G.

(2006). A 6-item scale for overall, emotional, and social loneliness: Confirmatory tests on survey data. Research on Aging, 28(5), 582–598. https://doi.org/10.1177/0164027506289723

14.

Dokkedahl, S. B., & Elklit, A. (2019). Understanding the mutual partner dynamic of intimate partner violence: A review. Partner Abuse, 10(3), 298-335. https://doi.org/10.1891/1946-6560.10.3.298

15.

Dokkedahl

Kok

R. N.

Murphy

Kristensen

T. R.

Bech-Hansen

Elklit

(2019). The psychological subtype of intimate partner violence and its effect on mental health: Protocol for a systematic review and meta-analysis. Systematic Reviews, 8, article 198. https://doi.org/10.1186/s13643-019-1118-1

16.

Dokkedahl

S. B.

Kirubakaran

Bech-Hansen

Kristensen

T. R.

Elklit

(2022). The psychological subtype of intimate partner violence and its effect on mental health: A systematic review with meta-analyses. Systematic Reviews, 11(1), 1–16. https://doi.org/10.1186/s13643-022-02025-z

17.

Doroszewicz

Forbes

G. B.

(2008). Experiences with dating aggression and sexual coercion among Polish college students. Journal of Interpersonal Violence, 23(1), 58–73. https://doi.org/10.1177/0886260507307651

18.

Eysenck

S. B.

Eysenck

H. J.

Barrett

(1985). A revised version of the psychoticism scale. Personality and Individual Differences, 6(1), 21–29. https://doi.org/10.1016/0191-8869(85)90026-1

19.

Gilbar

Charak

Trujillo

Cantu

J. I.

Cavazos

Lavi

(2023). Meta-analysis of cyber intimate partner violence perpetration and victimization: Different types and their associations with face-to-face IPV among men and women. Trauma, Violence, & Abuse, 24(3), 1948–1965. https://doi.org/10.1177/15248380221082087

20.

Grest

C. V.

Lee

J. O.

Gilreath

Unger

J. B.

(2018). Latent class analysis of intimate partner violence perpetration and victimization among latino emerging adults. Journal of Youth and Adolescence, 47, 575–585. https://doi.org/10.1007/s10964-017-0807-0

21.

Grygiel

Humenny

Rebisz

Świtaj

Sikorska

(2013). Validating the Polish adaptation of the 11-item de jong gierveld loneliness scale. European Journal of Psychological Assessment, 29(2), 129–139. https://doi.org/10.1027/1015-5759/a000130

22.

Hall

Hill

Moreland

Hales

G. K.

Boduszek

Debowska

(2023). Profiles of intimate partner violence victimization: A systematic review. Trauma, Violence, & Abuse, 24(5), 3280–3296. https://doi.org/10.1177/15248380221126183

23.

Hays

R. D.

Spritzer

K. L.

Thompson

W. W.

Cella

(2015). U.S. General population estimate for “excellent” to “poor” self-rated health item. Journal of General Internal Medicine, 30(10), 1511–1516. https://doi.org/10.1007/s11606-015-3290-x

24.

Hines

D. A.

Douglas

E. M.

(2009). Women’s use of intimate partner violence against men: Prevalence, implications, and consequences. Journal of Aggression, Maltreatment & Trauma, 18(6), 572–586. https://doi.org/10.1080/10926770903103099

25.

Hipp

J. R.

Bauer

D. J.

(2006). Local solutions in the estimation of growth mixture models. Psychological Methods, 11(1), 36–53. https://doi.org/10.1037/1082-989X.11.1.36

26.

Jaureguizar

Dosil-Santamaria

Redondo

Wachs

Machimbarrena

J. M.

(2024). Online and offline dating violence: Same same, but different? Psicologia: Reflexão e Crítica, 37(13), 1–17. https://doi.org/10.1186/s41155-024-00293-3

27.

Johnson

M. P.

(2006). Conflict and control: Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12(11), 1003–1018. https://doi.org/10.1177/1077801206293328

28.

Kroenke

Spitzer

R. L.

Williams

J. B.

Löwe

(2009). An ultra-brief screening scale for anxiety and depression: The PHQ–4. Psychosomatics, 50(6), 613–621. https://doi.org/10.1016/S0033-3182(09)70864-3

29.

Lagdon

Ross

Waterhouse-Bradley

Armour

(2023). Exploring the existence of distinct subclasses of intimate partner violence experience and associations with mental health. Journal of Family Violence, 38, 735–746. https://doi.org/10.1007/s10896-022-00393-7

30.

Lascorz

Larrañaga

Yubero

(2018). Prevalence of covert violence in intimate partner relationships. A study with Spanish university students. Open Journal of Social Sciences, 6(12), 37–53. https://doi.org/10.4236/jss.2018.612004

31.

Mendell

N. R.

Rubin

D. B.

(2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767–778. https://doi.org/10.1093/biomet/88.3.767

32.

Łukasik

Karakuła-Juchnowicz

Morylowska-Topolska

Flis

Krukow

(2015). Long-term somatic consequences of intimate partner violence in primary care female patients. Polski Merkuriusz Lekarski: Organ Polskiego Towarzystwa Lekarskiego, 39(234), 372–376. PMID: 26802690.

33.

Mandal

Gawor

Buczny

(2012). The stereotypes of man and woman in Poland—Content and factor structures. In Mandal

(Ed.), Masculinity and femininity in everyday life (pp. 11–32). Wydawnictwo Uniwersytetu Śląskiego.

34.

Minto

Masser

Louis

(2022). Lay understandings of the structure of intimate partner violence in relationships: An analysis of behavioral clustering patterns. Journal of Interpersonal Violence, 37(13-14), NP10810–NP10831. https://doi.org/10.1177/0886260520986276

35.

Murphy

C. M.

Hoover

S. A.

(1999). Measuring emotional abuse in dating relationships as a multifactorial construct. Violence and Victims, 14(1), 39–53. https://doi.org/10.1891/08866708.14.1.39

36.

Muthén

L. K.

Muthén

B. O.

(1998–2015). Mplus user’s guide (7^th ed.). Muthén & Muthén.

37.

Okuda

Olfson

Wang

Rubio

J. M.

Blanco

(2015). Correlates of intimate partner violence perpetration: Results from a national epidemiologic survey. Journal of Traumatic Stress, 28(1), 49–56. https://doi.org/10.1002/jts.21986

38.

O’Leary

K. D.

Slep

A. M. S.

(2012). Prevention of partner violence by focusing on behaviors of both young males and females. Prevention Science, 13(4), 329–339. https://doi.org/10.1007/s11121-011-0237-2

39.

Ortega Pacheco

Y. J.

Gallo-Barrera

Y. D.

Pozo Muñoz

(2024). Risk and protective factors for revictimization from intimate partner violence: A systematic review. International Journal of Social Psychiatry, 70(8), 1379–1391. https://doi.org/10.1177/00207640241270816

40.

Pico-Alfonso

M. A.

(2005). Psychological intimate partner violence: The major predictor of posttraumatic stress disorder in abused women. Neuroscience & Biobehavioral Reviews, 29(1), 181–193. https://doi.org/10.1016/j.neubiorev.2004.08.010

41.

Ramaswamy

DeSarbo

W. S.

Reibstein

D. J.

Robinson

W. T.

(1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Marketing Science, 12(1), 103–124. https://doi.org/10.1287/mksc.12.1.103

42.

Reichel

(2017). Determinants of intimate partner violence in Europe: The role of socioeconomic status, inequality, and partner behavior. Journal of Interpersonal Violence, 32(12), 1853–1873. https://doi.org/10.1177/0886260517698951

43.

Robins

R. W.

Hendin

H. M.

Trzesniewski

K. H.

(2001). Measuring global self-esteem: Construct validation of a single-item measure and the Rosenberg Self-Esteem Scale. Personality and Social Psychology Bulletin, 27(2), 151–161. https://doi.org/10.1177/0146167201272002

44.

Sardinha

Maheu-Giroux

Stöckl

Meyer

S. R.

García-Moreno

(2022). Global, regional, and national prevalence estimates of physical or sexual, or both, intimate partner violence against women in 2018. The Lancet, 399(10327), 803–813. https://doi.org/10.1016/S0140-6736(21)02664-7

45.

Schober

Boer

Schwarte

L. A.

(2018). Correlation coefficients: Appropriate use and interpretation. Anesthesia & Analgesia, 126(5), 1763–1768. https://doi.org/10.1213/ANE.0000000000002864

46.

Schokkenbroek

J. M.

Ouytsel

J. V.

Hardyns

Ponnet

(2022). Adults’ online and offline psychological intimate partner violence experiences. Journal of Interpersonal Violence, 37(15-16), NP14656–NP14671. https://doi.org/10.1177/08862605211015217

47.

Schwarz

(1978). Estimating the dimension of a model. The Annals of Statistics, 6(2), 461–464. https://doi.org/10.1214/aos/1176344136

48.

Sclove

S. L.

(1987). Application of model-selection criteria to some problems in multivariate analysis. Psychometrika, 52(3), 333–343. https://doi.org/10.1007/BF02294360

49.

Spencer

C. M.

Keilholtz

B. M.

Palmer

Vail

S. L.

(2024). Mental and physical health correlates for emotional intimate partner violence perpetration and victimization: A meta-analysis. Trauma, Violence, & Abuse, 25(1), 41–53. https://doi.org/10.1177/15248380221137686

50.

Ulloa

E. C.

Hammett

J. F.

(2016). The effect of gender and perpetrator–victim role on mental health outcomes and risk behaviors associated with intimate partner violence. Journal of Interpersonal Violence, 31(7), 1184–1207. https://doi.org/10.1177/0886260514564163

51.

Visschers

Jaspaert

Vervaeke

(2017). Social desirability in intimate partner violence and relationship satisfaction reports: An exploratory analysis. Journal of Interpersonal Violence, 32(9), 1401–1420. https://doi.org/10.1177/0886260515588922

52.

Wagers

S. M.

Piquero

A. R.

Narvey

Reid

J. A.

Loughran

T. A.

(2021). Variation in exposure to violence in early adolescence distinguishes between intimate partner violence victimization and perpetration among young men involved in the justice system. Journal of Family Violence, 36, 99–108. https://doi.org/10.1007/s10896-020-00170-4

53.

Wenger

M. R.

(2012). Patterns of misreporting intimate partner violence using matched pairs [Master’s thesis, The Pennsylvania State University]. PennState University Libraries. https://etda.libraries.psu.edu/catalog/15395.

54.

Williams

J. R.

Ghandour

R. M.

Kub

J. E.

(2008). Female perpetration of violence in heterosexual intimate relationships: Adolescence through adulthood. Trauma, Violence, & Abuse, 9(4), 227–249. https://doi.org/10.1177/1524838008324418

55.

Wolford-Clevenger

Zapor

Brasfield

Febres

Elmquist

Brem

, … Stuart

G. L.

(2016). An examination of the partner cyber abuse questionnaire in a college student sample. Psychology of Violence, 6(1), 156–162. https://doi.org/10.1037/a0039442

56.

World Health Organization (WHO). (2013). Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non- partner sexual violence. http://apps.who.int/iris/bitstream/10665/85239/1/9789241564625_eng.pdf.