The relationship between childhood trauma, doomscrolling, and dating violence among university students in Türkiye Authors: Vesile Adıgüzel, Anıl Eymur, Aslıhan Topcu Published: 2026 ABSTRACT BackgroundChildhood trauma, dating violence, and doomscrolling are increasingly drawing attention as important factors associated with young adults’ psychological well-being and interpersonal functioning. This study was conducted to examine the relationships among childhood trauma, doomscrolling, and dating violence among university students in Türkiye, and to evaluate the role of dating violence in the association between childhood trauma and doomscrolling.MethodThis descriptive, cross-sectional, correlational study was conducted with 502 university students in Türkiye between September 20 and November 20, 2025. Data were collected using the Personal Information Form, the Childhood Trauma Questionnaire (CTQ-28), the Dating Violence Scale (DVS), and the Doomscrolling Scale–Turkish Form (DS-TF). Number, percentage, mean, and standard deviation were used in data analysis; Pearson correlation analysis and simple linear regression analysis were used to examine the relationships among variables. PROCESS Model 4 was used for the mediation analysis. Ethics committee approval and institutional permission were obtained prior to the study.ResultsIt was determined that 72.7% of the participants were female and 34.9% were first-year students. The mean total scores were 43.66 ± 15.32 for CTQ-28, 45.11 ± 18.28 for DVS, and 33.94 ± 18.39 for DS-TF. Statistically significant positive correlations were found between CTQ-28 and DVS (r = 0.571, p ≤ 0.001), between CTQ-28 and DS-TF (r = 0.454, p ≤ 0.001), and between DVS and DS-TF (r = 0.478, p ≤ 0.001). Regression analyses showed that childhood trauma significantly predicted dating violence (F = 241.325, p ≤ 0.001; R² = 0.326), dating violence significantly predicted doomscrolling (F = 147.698, p ≤ 0.001; R² = 0.228), and childhood trauma significantly predicted doomscrolling (F = 129.861, p ≤ 0.001; R² = 0.206). Mediation analysis revealed that dating violence partially mediated the relationship between childhood trauma and doomscrolling.ConclusionPositive associations were found among childhood trauma, dating violence, and doomscrolling among university students in Türkiye. The findings support a significant indirect pathway through dating violence; however, causal interpretations are precluded by the cross-sectional design. Introduction Violence is a major public health problem that threatens individuals’ physical and psychological health and has serious consequences for the overall well-being of societies [1]. The World Health Organization defines violence as the intentional use of physical force or power, threatened or actual, against oneself, another person, a group, or a community, which results in or has a high likelihood of resulting in injury, death, psychological harm, developmental impairment, deprivation, or restriction [1]. According to the WHO classification, violence is addressed under three main categories: self-directed violence, interpersonal violence, and collective violence. Interpersonal violence includes family and intimate partner violence as well as community violence [1]. Within this framework, dating violence is an important form of interpersonal violence that includes physical, sexual, psychological, economic, and digital/virtual violent behaviors occurring between partners in a romantic relationship [2]. Dating violence encompasses a broad pattern of behaviors, including physical assault, sexual coercion, psychological abuse, controlling behaviors, threats, verbal harassment, covert monitoring, and stalking [2]. It has also been emphasized that this type of violence may emerge from the beginning of a relationship and continue recurrently across different stages of the relationship [2]. Reported to be prevalent across different societies and cultures, dating violence constitutes an important area of risk, particularly for young adults and university students [3, 4]. The emergence of dating violence is a multidimensional process shaped by the interaction of individual, familial, social, and environmental factors. Variables such as demographic characteristics, social position, family dynamics, socioeconomic status, and education have been reported to be associated with dating violence [5]. However, early life experiences have drawn particular attention in understanding this pattern of risk. Witnessing violence in the family environment, experiencing maltreatment during childhood, and being exposed to violent experiences in one’s surroundings are considered important risk domains that may influence later patterns of aggression or victimization [6]. Indeed, exposure to childhood maltreatment and traumatic experiences has been reported to be associated with an increased risk of aggression and violence-related behaviors later in life [7, 8]. For this reason, childhood trauma necessitates an understanding of dating violence not only in terms of current relationship dynamics but also within the context of the individual’s developmental history and relational patterns. According to attachment theory, early caregiving experiences shape individuals’ self-perceptions, expectations of others, and the emotional and behavioral responses they develop in close relationships. While secure, consistent, and responsive caregiving experiences support the development of secure attachment patterns, neglectful, abusive, frightening, or inconsistent caregiving may lay the groundwork for insecure attachment patterns. In this context, traumatic childhood experiences may be associated with relational insecurity, abandonment sensitivity, difficulties establishing intimacy, excessive approval-seeking, emotional vulnerability, and dysfunctional emotion regulation patterns. These patterns may, in turn, create a relational context that increases the risk of conflict, control, jealousy, aggression, or exposure to violence in romantic relationships [9]. In addition, such traumas have been reported to impair functioning in social relationships, reduce interpersonal trust, and constitute a significant risk background for anxiety and depressive symptoms [10]. When attachment theory and the emotion regulation perspective are considered together, childhood trauma appears to be associated with heightened threat perception, negative cognitive schemas, and dysfunctional coping patterns in later life [11]. Therefore, childhood trauma is considered a multidimensional risk factor linked not only to psychological symptoms but also to behavioral patterns that individuals develop in their romantic relationships [12–14]. The effects of early traumatic experiences may be observed not only in the domain of interpersonal relationships but also in individuals’ cognitive and behavioral patterns in digital environments. In particular, individuals with a trauma history may experience difficulties in emotion regulation, heightened threat perception, and sensitivity to uncertainty, which may increase their tendency to orient toward negative online content [15]. From an attachment-based perspective, insecure attachment patterns associated with early traumatic experiences may foster hypervigilance to threat and a biased attentional orientation toward negative or threatening cues. In digital environments characterized by a constant and rapidly updated flow of information, these tendencies may manifest as repetitive monitoring of distressing news content in an attempt to anticipate danger, reduce uncertainty, or regain a sense of control, thereby increasing vulnerability to doomscrolling. Within this context, “doomscrolling,” also referred to as compulsive consumption of negative news, has received increasing attention in recent years and is defined as persistently following negative news flow in online environments and devoting excessive time to it despite the negative affect it generates [16, 17]. The emergence and maintenance of doomscrolling may vary according to factors such as ways of coping with uncertainty, need for control, personality characteristics, and current psychological state [16]. Existing studies have shown that doomscrolling is associated with psychological distress and problematic social media use patterns, that intolerance of uncertainty may increase this behavior, and that psychological resilience may play a protective role [18–20]. The university period is an important component of the transition from adolescence to adulthood and has been conceptualized as a period of “emerging adulthood” [21]. During this period, some students leave their families for educational purposes and must adapt to a new academic system and social environment [22, 23]. The need to adapt across multiple domains, such as new responsibilities, independent living skills, changing teaching-learning expectations, and new friendship networks, may lead to marked changes in individuals’ educational, interpersonal, and social life patterns [23, 24]. It has been reported that academic and social adjustment demands during the transition to university may increase stress in some students and trigger maladaptive coping and problematic behavioral patterns [25]. At the same time, because romantic relationships intensify during this period, dating violence is emphasized as an important risk area for university students [5]. Studies showing that separation-individuation problems may be more pronounced in university students with adverse childhood experiences also support the potential effects of early experiences on university adjustment and relationship dynamics [26]. In addition, the intense interaction of university students with digital media makes this group an important population for studying problematic online behaviors such as doomscrolling. A review of the literature shows that studies examining childhood trauma, dating violence, and doomscrolling together within the same model among university students are limited. In particular, no study was found that examined whether dating violence is involved in an indirect association pattern between childhood trauma and doomscrolling. The aim of this study was to examine the relationships among childhood trauma, dating violence, and doomscrolling among university students in Türkiye. The study tested four hypotheses: (H1) a positive association between childhood trauma and dating violence; (H2) a positive association between childhood trauma and doomscrolling; (H3) a positive association between dating violence and doomscrolling; and (H4) a mediation model in which dating violence indirectly associates childhood trauma with doomscrolling. Method Research design This study was descriptive, cross-sectional, and correlational in nature. Sample and participants This study was conducted with students who were actively enrolled at a public university in Türkiye between September 20 and November 20, 2025, during the 2025–2026 academic year. The university was selected based on the researchers’ accessibility to participants and the feasibility of institutional collaboration. The study population consisted of all students who were actively enrolled at the university during the data collection period. The minimum required sample size was calculated based on the population size, a 95% confidence level, a 5% margin of error, and an assumed prevalence of 50% in order to provide the most conservative estimate when the prevalence was unknown. Based on this calculation, the minimum required sample size was determined to be 450. The final sample size was considered adequate for the planned analyses. Convenience sampling was used in the study. The inclusion criteria were being actively enrolled at the university during the 2025–2026 academic year, being 18 years of age or older, and providing informed consent for voluntary participation. The exclusion criteria were being under 18 years of age, refusing to participate in the study, and completing the questionnaire form incompletely. A total of 560 students were approached during the data collection process. Of these, 45 students declined to participate, and 13 students were excluded prior to analysis because their questionnaire forms contained incomplete data. Thus, the study was completed with a final sample of 502 students. This corresponds to an approximate response rate of 89.6%. Data collection The study data were collected through face-to-face interviews with students at the institution from which written permission had been obtained during the study period. The data were collected by the researchers (AE, AT). Each interview lasted approximately 15–20 min. Data collection instruments Study data were collected using a 21-item Personal Information Form, the Childhood Trauma Questionnaire (CTQ-28), the Dating Violence Scale (DVS), and the Doomscrolling Scale–Turkish Form (DS-TF). Personal information form Developed by the researchers, this form consists of 21 items designed to obtain participants’ personal information such as age, sex, year of study, and smoking status [27]. Childhood Trauma Questionnaire (CTQ-28) The Childhood Trauma Questionnaire, developed by Bernstein and colleagues to assess experiences of abuse and neglect before the age of 20, was adapted into Turkish and evaluated for validity and reliability by Şar, Öztürk, and İkikardeş (2012) [28]. The CTQ-28 is a 28-item, five-point Likert-type scale that assesses childhood trauma across five subscales. Response options for each item range from “1” (never) to “5” (very often): 1 = never, 2 = rarely, 3 = sometimes, 4 = often, and 5 = very often. In the original Turkish validation study, the Cronbach’s alpha coefficient was 0.93 [28]. In the present study, Cronbach’s alpha was 0.92. Dating Violence Scale (DVS) The Dating Violence Scale, developed by Gökkaya (2022) to measure dating violence behaviors in Turkish culture, is a 28-item, five-point Likert-type scale with four subscales. Items are rated as follows: 1 = never, 2 = rarely, 3 = sometimes, 4 = often, and 5 = very often. Total scores range from 28 to 140, with higher scores indicating higher levels of violent behavior. The Cronbach’s alpha coefficient for the total scale was reported as 0.89 [29]. In the present study, Cronbach’s alpha was 0.95. Doomscrolling Scale–Turkish Form (DS-TF) The original version of the scale was developed by Sharma et al. [17] to measure social media users’ doomscrolling behavior, and its validity and reliability for the Turkish population were established by Satici et al. [18]. The Doomscrolling Scale was found to be a reliable and valid measurement tool. It is a one-dimensional, 15-item, 7-point Likert-type scale ranging from 1 (“Strongly disagree”) to 7 (“Strongly agree”). The scale does not contain any reverse-coded items. The total score is used in the evaluation of the scale, and higher scores indicate higher levels of doomscrolling. In the reliability analysis of the scale, the internal consistency coefficient was reported as α = 0.938–0.944 [18]. In the present study, the Cronbach’s alpha coefficient of the scale was 0.94. Data analysis The data obtained in the study were analyzed using DataBeeg 1.0. Descriptive statistical methods were used to evaluate the data, and the variables were summarized using frequencies, percentages, minimum–maximum values, means, and standard deviations. Before conducting the parametric analyses, the normality of the data was assessed visually using Q-Q plots and statistically using skewness and kurtosis values. Skewness and kurtosis values within the range of ± 3 were considered acceptable for normal distribution [30]. Thirteen questionnaires with incomplete responses were excluded prior to analysis. Accordingly, all statistical analyses were conducted using complete-case data from the final sample of 502 participants. Since the analyzed dataset contained no missing values, no imputation procedure was applied. Pearson correlation analysis was performed to examine the associations among numerical variables. Simple linear regression analysis was conducted to examine the predictive relationships between the independent and dependent variables. In line with the theoretical model established in the study, the hypotheses were tested using Hayes’ PROCESS Macro. A simple mediation model (Model 4) was applied, and the significance of indirect effects was evaluated on the basis of 95% bias-corrected confidence intervals calculated using 5,000 bootstrap samples. In all statistical analyses, the level of significance was accepted as p < 0.05. Ethical considerations The study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval was obtained from the Non-Interventional Research Ethics Committee of a public university in Türkiye (27.11.2024/295), and written institutional permission was obtained from the study site (E-76051854-900-2600001801). Participants were informed about the study’s aim and invited to participate. Participation was ensured on a voluntary basis using an informed consent form in accordance with the Declaration of Helsinki principles, and both verbal and written informed consent were obtained from all students. Results The distribution of students according to their demographic characteristics is presented in Table 1. These variables were collected to characterize the study sample. Of the students, 50.0% were younger than 21 years, 72.7% were female, and 96.4% were single. It was determined that 34.9% were first-year students, 34.9% lived in a dormitory, 77.9% lived in a nuclear family, and 63.7% had 1–3 siblings. Regarding parental education, 29.3% of mothers and 30.7% of fathers were high school graduates. With respect to romantic relationship status, 40.8% reported, “I do not currently have a relationship, but I have had one in the past.” When daily internet use was examined, 54.8% reported using the internet for 4–6 h per day. In terms of social media platforms used, 85.3% reported using Instagram, 59.0% YouTube, 29.5% Twitter, 3.4% Facebook, 34.5% TikTok, 23.5% Snapchat, and 4.6% other platforms. Regarding smoking status, 67.7% reported not smoking. Regarding alcohol use, 79.3% reported no alcohol use. When the presence of a diagnosed psychological disorder was examined, 94.4% reported none. Finally, with respect to current psychotherapy use, 4.6% reported receiving psychotherapy, whereas 95.4% reported not receiving psychotherapy. Table 1Demographic Characteristics of the ParticipantsVariables n %Age(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{X}$$\end{document} ± SS, 20.87 ± 2.86)Under 2125150.021 years and older25150.0GenderWomen36572.7Men13727.3Marital statusMarried183.6Single48496.4ClassPreparatory class30.61. Class17534.92. Class10821.53. Class13426.74. Class8216.3Place of residenceDormitory35370.3Student house6412.7My family8416.7Relatives10.3Family typeNuclear family39177.9Extended family8216.3Divorced/broken family295.8Number of siblingsNo224.41–332063.74–612023.97–9346.810 and above61.2Mother’s educational statusElementary School19238.2Middle School10520.9High School14729.3University5410.8Graduate/Doctorate40.8Father’s educational statusElementary School12023.9Middle School10420.7High School15430.7University10921.7Graduate/Doctorate153.0The best-expressed attitude toward romantic relationshipsI’m currently in a relationship.17534.9I don’t have a relationship right now, but I have had one before.20540.8I haven’t had a relationship so far.12224.3Internet usage time during the day1–3 h8516.94–6 h27554.87–9 h11422.710–12 h204.012 h and over81.6Social media usedInstagramYes42885.3No7414.7YouTubeYes29659.0No20641.0TwitterYes14829.5No35470.5FacebookYes173.4No48596.6Tik TokYes17334.5No32965.5SnapchatYes11823.5No38476.5OtherYes234.6No47995.4SmokingYes13927.7No34067.7I quit234.6Alcohol consumptionYes8216.3No39879.3I quit224.4The presence of a recognized psychological disorderYes285.6No47494.4Current status of receiving psychotherapy servicesYes234.6No47995.4Total502100.0 Table 1Demographic Characteristics of the ParticipantsVariables n %Age(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{X}$$\end{document} ± SS, 20.87 ± 2.86)Under 2125150.021 years and older25150.0GenderWomen36572.7Men13727.3Marital statusMarried183.6Single48496.4ClassPreparatory class30.61. Class17534.92. Class10821.53. Class13426.74. Class8216.3Place of residenceDormitory35370.3Student house6412.7My family8416.7Relatives10.3Family typeNuclear family39177.9Extended family8216.3Divorced/broken family295.8Number of siblingsNo224.41–332063.74–612023.97–9346.810 and above61.2Mother’s educational statusElementary School19238.2Middle School10520.9High School14729.3University5410.8Graduate/Doctorate40.8Father’s educational statusElementary School12023.9Middle School10420.7High School15430.7University10921.7Graduate/Doctorate153.0The best-expressed attitude toward romantic relationshipsI’m currently in a relationship.17534.9I don’t have a relationship right now, but I have had one before.20540.8I haven’t had a relationship so far.12224.3Internet usage time during the day1–3 h8516.94–6 h27554.87–9 h11422.710–12 h204.012 h and over81.6Social media usedInstagramYes42885.3No7414.7YouTubeYes29659.0No20641.0TwitterYes14829.5No35470.5FacebookYes173.4No48596.6Tik TokYes17334.5No32965.5SnapchatYes11823.5No38476.5OtherYes234.6No47995.4SmokingYes13927.7No34067.7I quit234.6Alcohol consumptionYes8216.3No39879.3I quit224.4The presence of a recognized psychological disorderYes285.6No47494.4Current status of receiving psychotherapy servicesYes234.6No47995.4Total502100.0 Demographic Characteristics of the Participants Age (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{X}$$\end{document} ± SS, 20.87 ± 2.86) The mean total CTQ-28 score of the participating students was 43.66 ± 15.32, the mean total DVS score was 45.11 ± 18.28, and the mean total DS-TF score was 33.94 ± 18.39 (Table 2). Table 2Mean Total Scores of the ScalesScalesMin- MaxMean ± SDChildhood Trauma Questionnaire25.00–85.0043.66 ± 15.32Dating Violence Scale28.00-109.0045.11 ± 18.28Doomscrolling Scale–Turkish Form15.00-101.0033.94 ± 18.39Min Minimum, Max Maximum, SD Standard Deviation Table 2Mean Total Scores of the ScalesScalesMin- MaxMean ± SDChildhood Trauma Questionnaire25.00–85.0043.66 ± 15.32Dating Violence Scale28.00-109.0045.11 ± 18.28Doomscrolling Scale–Turkish Form15.00-101.0033.94 ± 18.39Min Minimum, Max Maximum, SD Standard Deviation Mean Total Scores of the Scales Min Minimum, Max Maximum, SD Standard Deviation CTQ-28 was positively and significantly associated with both DVS (r = 0.571, p ≤ 0.001) and DS-TF (r = 0.454, p ≤ 0.001). In addition, a statistically significant and positive association was found between DVS and DS-TF (r = 0.478, p ≤ 0.001) (Table 3). Table 3Correlations Among the ScalesScales1231- Childhood Trauma Questionnairer1.0000.5710.454p- ≤ 0.001* ≤ 0.001* 2- Dating Violence Scaler1.0000.478p- ≤ 0.001* 3- Doomscrolling Scale–Turkish Formr1.000p-Pearson correlation coefficients are presented. All reported p-values were ≤ 0.001 Table 3Correlations Among the ScalesScales1231- Childhood Trauma Questionnairer1.0000.5710.454p- ≤ 0.001* ≤ 0.001* 2- Dating Violence Scaler1.0000.478p- ≤ 0.001* 3- Doomscrolling Scale–Turkish Formr1.000p-Pearson correlation coefficients are presented. All reported p-values were ≤ 0.001 Correlations Among the Scales Pearson correlation coefficients are presented. All reported p-values were ≤ 0.001 Table 4 presents the results of the regression analyses. The model established to examine whether childhood trauma predicted dating violence was found to be statistically significant (F = 241.325, p ≤ 0.001). The findings indicated that childhood trauma significantly and positively predicted dating violence (β = 0.681, p ≤ 0.001). The model explained 32.6% of the variance in dating violence. Table 4Regression Analysis ResultsβStandard ErrorBetat p 1 Dating Violence Scale Childhood Trauma Questionnaire 0.6810.0440.57115.535 ≤ 0.001* 2 Doomscrolling Scale–Turkish Form Dating Violence Scale 0.4810.0400.47812.153 ≤ 0.001* 3 Doomscrolling Scale–Turkish Form Childhood Trauma Questionnaire 0.5450.0480.45411.396 ≤ 0.001* p ≤ 0.001* Table 4Regression Analysis ResultsβStandard ErrorBetat p 1 Dating Violence Scale Childhood Trauma Questionnaire 0.6810.0440.57115.535 ≤ 0.001* 2 Doomscrolling Scale–Turkish Form Dating Violence Scale 0.4810.0400.47812.153 ≤ 0.001* 3 Doomscrolling Scale–Turkish Form Childhood Trauma Questionnaire 0.5450.0480.45411.396 ≤ 0.001* p ≤ 0.001* Regression Analysis Results p ≤ 0.001* The model established to examine whether dating violence predicted doomscrolling was also statistically significant (F = 147.698, p ≤ 0.001). The findings indicated that dating violence significantly and positively predicted doomscrolling (β = 0.481, p ≤ 0.001). The model explained 22.8% of the variance in doomscrolling. The model established to examine whether childhood trauma predicted doomscrolling was found to be statistically significant (F = 129.861, p ≤ 0.001). The findings indicated that childhood trauma significantly and positively predicted doomscrolling (β = 0.545, p ≤ 0.001). The model explained 20.6% of the variance in doomscrolling. A model was constructed to examine the mediating role of dating violence in the relationship between childhood trauma and doomscrolling. First, the total effect of childhood trauma on doomscrolling was evaluated. The analysis showed that childhood trauma significantly and positively predicted doomscrolling (β = 0.5453, p ≤ 0.001). Next, the indirect effect of childhood trauma on doomscrolling through dating violence was examined, and this indirect effect was found to be statistically significant because the bootstrap confidence interval did not include zero (95% CI = 0.1393–0.3029). Finally, when dating violence was included in the model, the direct effect of childhood trauma on doomscrolling remained statistically significant but decreased in magnitude (β = 0.3233, p ≤ 0.001). These results indicate a partial mediation model, in which dating violence significantly mediated the relationship between childhood trauma and doomscrolling (Table 5) (Fig. 1). Table 5Mediation Analysis ResultsEffectβStandard Errort p SignificantChildhood Trauma Questionnaire → Doomscrolling Scale–Turkish Form0.54530.047811.3956 ≤ 0.001* AcceptChildhood Trauma Questionnaire → Dating Violence Scale → Doomscrolling Scale–Turkish FormDirect Effect0.32330.05575.8091 ≤ 0.001* AcceptIndirect Effect0.2219Confidence Interval(0.1393, 0.3029)Meaningful*p ≤ 0.001 Table 5Mediation Analysis ResultsEffectβStandard Errort p SignificantChildhood Trauma Questionnaire → Doomscrolling Scale–Turkish Form0.54530.047811.3956 ≤ 0.001* AcceptChildhood Trauma Questionnaire → Dating Violence Scale → Doomscrolling Scale–Turkish FormDirect Effect0.32330.05575.8091 ≤ 0.001* AcceptIndirect Effect0.2219Confidence Interval(0.1393, 0.3029)Meaningful*p ≤ 0.001 Mediation Analysis Results Confidence Interval (0.1393, 0.3029) *p ≤ 0.001 Fig. 1Mediation between the scales Mediation between the scales Discussion This study demonstrated statistically significant positive relationships among childhood trauma, dating violence, and doomscrolling. The findings indicate that higher levels of childhood trauma were related to higher levels of dating violence and doomscrolling, and that dating violence was also positively related to doomscrolling. These results support the relevance of early adverse experiences to later relational and digital behavioral patterns among university students. However, because of the cross-sectional design, the findings should be interpreted within a correlational framework and do not permit causal or temporal inferences. A moderate-to-strong positive correlation was found between CTQ-28 and DVS scores (r = 0.571, p ≤ 0.001). This finding aligns with previous literature showing a positive relationship between adverse childhood experiences and dating violence or intimate partner violence victimization [14, 31]. Similarly, Cho et al. (2024) reported that childhood maltreatment was positively associated with dating violence victimization among young people [12]. Wolford-Clevenger et al. (2016) also emphasized that childhood trauma is linked to interpersonal vulnerability through coping processes and decision-making patterns [32]. In the present study, higher childhood trauma scores were associated with higher dating violence scores. However, because of the cross-sectional design, this relationship should not be interpreted as causal or temporally ordered. A moderate positive correlation was also found between CTQ-28 and doomscrolling (r = 0.454, p ≤ 0.001). This finding is consistent with literature indicating that early adverse experiences are linked to cognitive-emotional processes such as threat-focused attention, hypervigilance, intolerance of uncertainty, and a tendency toward negative information seeking [18, 19, 33, 34]. Evidence showing that doomscrolling is positively associated with depression, future anxiety, and psychological distress, and negatively associated with psychological well-being, further supports this interpretation [35]. In the present study, higher childhood trauma scores were associated with higher doomscrolling scores. However, the cross-sectional nature of the data does not allow the conclusion that childhood trauma increases doomscrolling behavior, nor does it establish temporal directionality. A moderate positive correlation was found between DVS and DS-TF scores (r = 0.478, p ≤ 0.001). This finding indicates that higher levels of dating violence were associated with higher levels of doomscrolling. Although studies directly examining these two variables together remain limited, this result suggests that dating violence may be relevant to problematic negative information consumption in digital environments. Possible explanations include heightened psychological distress, increased threat sensitivity, or a stronger attentional orientation toward negative content among individuals exposed to dating violence [35]. However, these mechanisms were not directly tested in the present study. Therefore, while the finding supports a significant relationship between dating violence and doomscrolling, it does not establish causal direction or underlying explanatory pathways. The regression analyses also supported the correlational findings. The model examining whether childhood trauma predicted dating violence was statistically significant and explained 32.6% of the variance. This proportion indicates that childhood trauma was a meaningful statistical predictor of dating violence in this sample. Similarly, the model examining whether childhood trauma predicted doomscrolling was significant and explained 20.6% of the variance. The model examining whether dating violence predicted doomscrolling explained 22.8% of the variance. These values indicate that the relationships among the variables were statistically significant and showed moderate explanatory power; however, they also indicate that a multifactorial behavior such as doomscrolling cannot be explained by a single variable alone. The mediation analysis findings in this study support the hypothesized indirect pathway among childhood trauma, dating violence, and doomscrolling. Specifically, the results indicated a partial mediation model in which dating violence significantly mediated the relationship between childhood trauma and doomscrolling. However, because the cross-sectional design does not permit the establishment of temporal ordering among variables, these findings should not be interpreted as evidence of a causal mediational mechanism. In addition, reverse or reciprocal relationships cannot be ruled out. Therefore, although the findings support the proposed indirect pathway, this model should be examined further in longitudinal studies and in studies testing alternative models. The mean CTQ-28 total score (43.66 ± 15.32) fell within a moderate range, indicating that childhood trauma was reported by a notable proportion of university students in this sample. This result is comparable to findings from previous studies conducted with university students [36, 37]. The mean DVS total score was low, indicating that reported dating violence levels were relatively limited in this sample; however, this should not be interpreted as indicating that violence is unimportant or that the participants necessarily have healthy relationship patterns. Similarly, the mean DS-TF total score was low, suggesting that doomscrolling levels were relatively limited at the overall sample level. Nevertheless, low mean levels do not mean that there are no significant associations among the variables. Limitations This study has several limitations. First, because the research data were obtained only from students actively enrolled at a public university in Türkiye, the generalizability of the findings is limited. In addition, although female participants constituted a large proportion of the sample, gender-based subgroup analyses were not conducted, which restricted the evaluation of possible gender differences. This is particularly important because the prevalence, expression, and dynamics of dating violence may differ by gender; therefore, the findings may primarily reflect female students’ experiences and should be generalized to male students with caution. Second, all variables were assessed using self-report measures. This may have increased the risk of social desirability bias, response bias, and common method variance. Because all study variables were collected from the same source at a single time point, the possibility of common method bias should be considered when interpreting the observed associations. Moreover, although the instruments used in this study have demonstrated acceptable psychometric properties in previous research, confirmatory factor analysis was not conducted in the present sample. Therefore, the factor structure and model fit of the scales were not re-examined in this sample. Third, the study employed a cross-sectional design. Therefore, the observed associations and indirect effects should not be interpreted as evidence of causality or temporal ordering. The possibility of reverse or reciprocal relationships among the variables cannot be ruled out. In this context, the findings derived from the mediation analysis should be interpreted as statistical indications of indirect associations rather than evidence of a causal mediational mechanism. In addition, alternative models to the proposed model were not tested. Fourth, demographic variables were not included as covariates in the analytical model; therefore, potential confounding effects cannot be completely ruled out. In addition, multicollinearity among the predictors was not separately assessed. Similarly, the variables measured through multi-item scales were analyzed using total scores, and Structural Equation Modeling, which would allow the simultaneous testing of relationships among latent constructs while accounting for measurement error, was not employed. Although several demographic variables were collected and descriptively presented, these variables were not included as covariates in the main analytical model. Therefore, their potential confounding effects could not be evaluated and should be addressed in future studies. In addition, although the CTQ-28 and DVS include theoretically meaningful subscales, the present study analyzed only total scores in order to focus on the overall associations among childhood trauma, dating violence, and doomscrolling within a parsimonious model. Therefore, potential differences related to specific trauma types or dimensions of dating violence could not be examined. Future studies may provide a more nuanced understanding by evaluating subscale-level associations. Finally, possible psychiatric covariates such as depression, anxiety, and stress were not measured in this study. This is an important limitation because these variables may function as potential confounders, given that they are known to be associated with childhood trauma and may also be related to both dating violence and doomscrolling. Therefore, interpretations regarding the psychological triggers and consequences associated with doomscrolling could not be directly tested with the present data and should be evaluated with caution. Conclusion This study found positive associations between childhood trauma, dating violence, and doomscrolling among university students. In addition, the findings indicated that dating violence significantly partially mediated the relationship between childhood trauma and doomscrolling. However, due to the cross-sectional and correlational design of the study, these findings should not be interpreted as evidence of causality or temporal ordering. The university years represent an important developmental period in which individuals experience the transition from adolescence to young adulthood and encounter new social, academic, and interpersonal adjustment demands. In this context, early adverse experiences may shape individuals’ stress-coping patterns, romantic relationship experiences, and forms of digital media use. Similarly, doomscrolling may reflect negative affect, threat-focused attention, or avoidance-based coping patterns in some students. However, these possible explanations were not directly tested in the present study. Campus mental health services should prioritize the early identification of students with histories of childhood trauma and dating violence. Based on these findings, psychoeducational and psychosocial interventions should incorporate components addressing healthy digital media use, coping with triggering content, emotion regulation, stress management, healthy relationship skills, recognition of violence, and help-seeking behaviors. Nevertheless, experimental, longitudinal, and intervention-based studies are needed to establish the effectiveness of such approaches.