Predicting probabilities of bullying victimization or perpetration in adolescents: a multinomial logistic regression analysis

<jats:p>&lt;p&gt;&lt;strong&gt;Context and relevance. &lt;/strong&gt;Bullying remains a pervasive challenge across educational systems worldwide, posing serious risks to students' psychological well-being and academic performance. In Indonesia, its prevalence continues to grow, echoing patterns observed in various other countries and reinforcing the need for culturally informed, data-driven interventions. &lt;strong&gt;Objective. &lt;/strong&gt;This study investigates high school students' risk of involvement in bullying &amp;mdash; either as victims or perpetrators &amp;mdash; based on age, gender, and school type. &lt;strong&gt;Methods and materials. &lt;/strong&gt;Utilizing a cross-sectional survey design, data were collected from 1153 students aged 14 to 19 years (M = 16; SD = 0,919), including 514 males (45%) and 639 females (55%). The sample comprised 907 senior high school students and 246 vocational high school students from 10 sub-districts in Medan City, Indonesia. &lt;strong&gt;Results. &lt;/strong&gt;A multinomial logistic regression analysis revealed that gender was the only significant predictor of victimization risk, while no variables significantly predicted dual roles as both perpetrator and victim. &lt;strong&gt;Conclusions. &lt;/strong&gt;These findings underscore the importance of gender-sensitive monitoring and preventive strategies, particularly for female students, and contribute to a growing body of international research aimed at mitigating bullying behaviors through targeted educational programs.&lt;/p&gt;</jats:p>