Examining Trajectories of Peer Violence Experiences and Their Association with Gender Norms and Agency Among Adolescents: Evidence from China, Democratic Republic of Congo, and Indonesia.
Astha Ramaiya, Mengmeng Li, Chunyan Yu, Aim��e Lulebo, Anggriyani W Pinandari, Jhumka Gupta, Michelle R Kaufman, Bushra Sabri, Shoshanna Fine, Sam Beckwith
Author Information
Astha Ramaiya: Johns Hopkins University, Baltimore, MD, USA. ORCID
Mengmeng Li: Johns Hopkins University, Baltimore, MD, USA.
Chunyan Yu: Institute of public health and health policy, Shanghai-MOST Key Laboratory of Health and Disease Genomics, NHC Key Lab of Reproduction Regulation, Shanghai Institute for Biomedical and Pharmaceutical Technologies, China.
Aim��e Lulebo: Kinshasa School of Public Health and College of Medicine, Democratic Republic of Congo.
Anggriyani W Pinandari: University of Gadjah Mada, Yogyakarta, Indonesia.
Jhumka Gupta: George Mason University, Fairfax, VA, USA.
Michelle R Kaufman: Johns Hopkins University, Baltimore, MD, USA. ORCID
Bushra Sabri: Johns Hopkins University, Baltimore, MD, USA. ORCID
Shoshanna Fine: Johns Hopkins University, Baltimore, MD, USA. ORCID
Sam Beckwith: Johns Hopkins University, Baltimore, MD, USA.
We examine peer violence trajectories among a purposive sample of urban poor adolescents over 3���years in Kinshasa, Democratic Republic of Congo (DRC); Shanghai, China, and Bandar Lampung, Denpasar, and Semarang, Indonesia. We examine peer violence experiences over time and seek to understand the relationship between gender norm perceptions and personal agency on peer violence experiences. A longitudinal (wave 1-3) panel was used from the three study sites outlined above (���=���6,660). Adolescents were purposefully selected if they were 10 to 14���years old at wave 1 and living in urban poor communities. A repeated measure latent class analysis was used to parameterize peer violence experiences over time. Next, we conducted multinomial logistic regression models to examine the relationship between gender norms perceptions and agency (personal attribute of empowerment) with peer violence class membership. A five-class model was best fitting across all sites with four consistent classes: "low peer violence" (experienced by most); "declining violence"; "peaks in middle"; and "consistently high violence." In the three cities in Indonesia and Kinshasa, DRC, there was an "increasing violence" class, and in Shanghai, China, there was a "high emotional victimization" class. In general, more endorsement of unequal gender norms and higher agency were associated with greater relative risk of any violence victimization and/or perpetration versus the low violence class membership. However, associations between gender norm perceptions and agency on peer violence trajectories differed within and between sites. These results compel programs to understand and address the underlying beliefs that promote peer violence. Longitudinal trends of chronic offenders extenuate the need to include family- and community-level interventions to prevent and mitigate the effects of peer violence. Additionally, there is a need to address social and structural determinants such as gender and power to promote lifelong health free of injury and violence.
