Who Gets Targeted Online? Study Maps Cybervictimization Among Iranian Students

Cyberbullying has long been framed as a problem of adolescence, a digital-era extension of the schoolyard shove. But a new study from Iran suggests that the experience of being harassed, humiliated, or threatened online does not stop when young people enter university. In one of the most detailed psychometric investigations of its kind in a non-Western higher-education setting, researchers report that cybervictimization among Iranian university students follows a distinct social pattern, one shaped more by gender and socioeconomic position than by age itself.

The study, published as an open-access research article in BMC Psychology, surveyed 828 students aged 17 to 25 enrolled across three major public universities in Iran. The participants had a mean age of 21.05 years with a standard deviation of 1.82, and roughly two-thirds of the sample, 67.6 percent, were female. Using a cross-sectional design, the researchers administered the 12-item Cyber-Bullying/Victimization Experiences Questionnaire, known as the CBVEQ, and then subjected the resulting data to a battery of statistical tests designed to answer two questions at once: how severe is cybervictimization in this population, and which social and demographic characteristics predict it?

The technical rigor of the measurement work is one of the study’s standout features. Rather than simply summing questionnaire responses and moving on, the team used confirmatory factor analysis, or CFA, with a Weighted Least Squares Means and Variance adjusted estimator, abbreviated WLSMV. This estimator is particularly well suited to ordered categorical questionnaire data, which are common in victimization scales where response options are graded rather than continuous. The result was striking: a unidimensional model, meaning that all twelve items tap a single underlying construct of cybervictimization, fit the data exceptionally well. The comparative fit index reached 0.995 and the Tucker-Lewis index 0.994, both far above conventional thresholds of 0.95, while the root mean square error of approximation stood at 0.075 and the standardized root mean square residual at 0.060. Standardized factor loadings across all items ranged from 0.64 to 0.88, indicating that every item contributed meaningfully to the measurement of the construct.

Perhaps more important for future research is what the team found when they tested the scale across genders. Using sequential multi-group confirmatory factor analysis, they demonstrated full scalar measurement invariance between male and female students. In practical terms, this means the questionnaire measures the same construct, in the same way, for both genders: a given score reflects the same level of victimization whether the respondent is a man or a woman. The fit changes were negligible, with a change in CFI of just +0.001 and a change in RMSEA of -0.003. Without such invariance, comparisons of average scores between genders can be misleading, because differences might reflect the instrument behaving differently across groups rather than genuine differences in experience. The confirmation of full scalar invariance gives researchers a validated license to make those comparisons in Iranian and potentially other comparable populations.

The distribution of victimization scores tells its own story. The mean total score on the CBVEQ was 16.45 with a standard deviation of 6.84, but the distribution was strongly floor-concentrated and right-skewed, with a skewness of 2.359 and a kurtosis of 5.877. In plain language, most students reported relatively low levels of victimization, while a smaller group carried the weight of the distribution with substantially higher scores. This pattern is typical of victimization research: harm online is not evenly spread but concentrated among a vulnerable minority. It also explains the authors’ methodological choices, since heavily skewed data of this kind can violate the assumptions of ordinary least squares approaches and call for robust estimation strategies.

When the researchers turned to predictors, they used multivariable linear regression, entering six socio-demographic variables simultaneously so that each estimate reflects the unique contribution of that factor while holding the others constant. The overall model was statistically significant, with F(6, 821) = 7.932 and p < 0.001, though the proportion of variance explained, R-squared of 0.055, was modest. That modesty is itself informative: socio-demographics alone capture only a small slice of why some students are victimized and others are not, pointing to the likely importance of unmeasured factors such as personality, online behavior patterns, peer networks, and platform-specific exposure.

Within that model, however, three predictors stood out. Male gender was the strongest, with a standardized coefficient of 0.193 and p < 0.001, indicating that men reported significantly higher victimization severity than women in this sample. Higher parental education also predicted elevated victimization, with a standardized beta of 0.105 and p = 0.003. Conversely, students who rated their family’s economic status more favorably reported lower victimization severity, with a standardized coefficient of -0.095 and p = 0.038. Notably, age, employment status, and family income adequacy showed no significant associations once the other variables were accounted for.

The gender finding runs counter to a substantial body of Western research, much of which has found that girls and young women experience higher rates of certain forms of cybervictimization, particularly those with sexualized or appearance-related content. The authors of the new study interpret their result through the lens of gendered interaction dynamics, suggesting that the ways in which men and women engage online, and the norms governing those interactions in the Iranian context, may shape who becomes a target. They caution against simplistic readings, but the finding underscores a broader point that has emerged repeatedly in cross-cultural cyberpsychology: the digital experience of harassment is not culturally uniform, and interventions imported wholesale from one context may misfire in another.

The socioeconomic pattern is equally thought-provoking. The association between higher parental education and greater victimization, alongside the protective effect of higher subjective economic status, hints at a complicated relationship between social stratification and online risk. One possibility the findings raise is that students from more educated households spend more time online or engage more intensively with digital platforms, increasing exposure. Subjective economic status, meanwhile, may operate through confidence, social capital, or access to supportive resources that buffer against harassment. The authors emphasize that cybervictimization severity in this population is tied to gendered interaction dynamics and socioeconomic status rather than chronological age, a conclusion with direct implications for how universities design digital-safety programs.

That implication is the study’s clearest practical takeaway. If risk profiles differ by gender and by social position, then one-size-fits-all awareness campaigns are unlikely to be efficient. The authors argue that interventions targeting digital safety in higher education should account for gender-specific risk profiles and social stratification factors, tailoring outreach to the students who are statistically most at risk rather than treating the entire student body as uniformly vulnerable. The concentration of victimization in a skewed tail of the distribution reinforces this: resources aimed at the highest-scoring minority are likely to reach the students who need them most.

The study also fills a genuine gap in the literature. The authors note that empirical investigations of the socio-demographic determinants of cybervictimization among emerging adults in higher education remain limited, particularly outside Western settings. Most of what is known about online victimization in young adulthood comes from North America and Europe, and instruments validated there do not automatically transfer across languages and cultures. By validating the CBVEQ in an Iranian university population, demonstrating its unidimensional structure, and establishing gender measurement invariance, the team has provided a psychometrically sound tool for future comparative work, both within Iran and across national boundaries.

There are, of course, limits to what a single cross-sectional survey can establish. The design captures a snapshot in time and cannot determine whether the measured characteristics cause victimization or merely correlate with it. The modest R-squared value makes clear that much of the variance in victimization lies outside the socio-demographic variables studied. And because the sample was drawn from public universities, generalization to other segments of Iranian emerging adults requires caution. Still, the study’s combination of a large sample, rigorous factor-analytic validation, and a multivariable predictive framework makes it a valuable data point in a rapidly growing field.

As universities worldwide grapple with how to protect students in digital spaces, this research offers a reminder that the answer depends on context. Who gets targeted online, and how severely, is shaped by the interplay of gender norms, family background, and economic perception, factors that vary across societies and that no universal template can fully anticipate. With a validated, gender-invariant measurement instrument now available for Iranian university populations, researchers and student-welfare professionals in the region have a firmer foundation on which to build that understanding, and to design interventions that meet the actual risk landscape rather than an imported assumption of it.

Subject of Research: Socio-demographic predictors and psychometric measurement of cybervictimization among Iranian university students

Article Title: Cybervictimization among Iranian university students

Article References: Zarei Nouroozi, M., Feyzi Dehkharghani, S., Sharifi, A., Nejatifar, S., Rezaei, S., & Goodarzi Ardakani, Z. (2026). Cybervictimization among Iranian university students. BMC Psychology. https://doi.org/10.1186/s40359-026-05738-1

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05738-1

Keywords: cybervictimization, cyberbullying, university students, Iran, emerging adulthood, psychometrics, measurement invariance, confirmatory factor analysis, gender differences, socioeconomic status, BMC Psychology, digital safety