Data_Sheet_5_A systematic review and meta-analysis of interventions to decrease cyberbullying perpetration and victimization: An in-depth analysis within the Asia Pacific region.PDF

BackgroundCyberbullying perpetration and victimization are prevalent issues in adolescent development and are a rising public health concern. Numerous interventions have been developed and implemented to decrease cyberbullying perpetration and victimization. Through an updated systematic review and meta-analysis, this study aimed to tackle a significant gap in the cyberbullying literature by addressing the need to empirically determine the effectiveness of programs with non-school-aged samples with a specific focus on studies conducted within the Asia-Pacific region.
MethodsA systematic literature review was conducted to identify intervention research to reduce cyberbullying perpetration and victimization published from January 1995 to February 2022. Ten electronic databases—Cambridge Journal Online, EBSCOHOST, ERIC, IEEE XPLORE, Oxford Journal Online, ProQuest Dissertations and Theses, PubMed (Medline), Science Direct, Scopus, Springerlink—and a subsequent manual search were conducted. Detailed information was extracted, including the summary data that could be used to estimate effect sizes. The studies’ methodological quality was assessed using the Effective Public Health Practice Project (EPHPP) quality assessment tool.
FindingsEleven studies were included in the review of the 2,540 studies identified through databases, and 114 additional records were discovered through citation searching. Only four studies were included in the meta-analysis, exploring game-based, skill-building, school-based, and whole-school interventions. The first meta-analysis pooled estimates from these four studies that assessed cyberbullying perpetration frequency using continuous data post-intervention. These studies reported data from 3,273 participants (intervention n = 1,802 and control n = 1,471). A small but not statistically significant improvement favoring the intervention group from pre- to post-intervention was shown by the pooled effect size, −0.04 (95% CI [−0.10,0.03], Z = 1.11, P = 0.27). The second meta-analysis included two qualified studies investigating cyberbullying victimization frequency using continuous data at post-intervention among 2,954 participants (intervention n = 1,623 and control n = 1,331). A very small but non-significant effect favoring the intervention group was discovered.
ConclusionThis research primarily highlights that the endeavor for cyberbullying intervention is still developing in the Asia-Pacific region, currently involving a limited set of stakeholders, settings, and delivery modes. Overall, meta-analyses of cyberbullying interventions conducted in the Asia Pacific found no significant effects in reducing cyberbullying perpetration and victimization.

研究背景：网络欺凌（Cyberbullying）施害行为与受害经历是青少年发展进程中的普遍问题，同时也是日益受到关注的公共卫生议题。目前已有大量干预方案被开发并实施，旨在降低网络欺凌施害与受害的发生风险。本研究通过开展更新版系统综述（systematic review）与元分析（meta-analysis），旨在填补网络欺凌研究领域的显著空白——学界亟需通过实证手段明确面向非学龄样本的干预项目的有效性，且本研究特别聚焦于亚太地区开展的相关研究。
研究方法：本研究开展系统综述，以筛选1995年1月至2022年2月期间发表的、旨在降低网络欺凌施害与受害行为的干预类研究。检索共覆盖10个电子数据库：剑桥期刊在线（Cambridge Journal Online）、EBSCOHOST、教育资源信息中心（ERIC, Education Resources Information Center）、IEEE XPLORE、牛津期刊在线（Oxford Journal Online）、ProQuest学位论文全文库（ProQuest Dissertations and Theses）、PubMed（Medline）、Science Direct、Scopus、Springerlink，随后辅以人工检索。研究团队提取了详细信息，包括可用于计算效应量（effect size）的汇总数据；并采用公共卫生实践效果项目（Effective Public Health Practice Project, EPHPP）质量评价工具，对纳入研究的方法学质量进行评估。
研究结果：本研究通过数据库检索共识别出2540项研究，经筛选后纳入11项研究；此外通过引文检索额外发现114条相关记录。最终仅有4项研究被纳入元分析，其干预类型涵盖基于游戏的干预、技能培养干预、基于学校的干预以及全校干预。第一项元分析合并了这4项研究的效应估计值，这些研究均采用干预后连续数据评估网络欺凌施害行为发生频率，共纳入3273名参与者（干预组n=1802，对照组n=1471）。合并效应量为-0.04（95%置信区间（confidence interval, CI）[−0.10, 0.03]，Z=1.11，P=0.27），结果显示干预组相较于对照组从基线至干预后仅存在小幅改善，但未达到统计学显著性水平。第二项元分析纳入2项符合标准的研究，采用干预后连续数据评估网络欺凌受害行为发生频率，共纳入2954名参与者（干预组n=1623，对照组n=1331）。结果显示干预组同样存在小幅获益，但未达到统计学显著性水平。
研究结论：本研究主要表明，亚太地区的网络欺凌干预工作仍处于发展阶段，目前涉及的利益相关方、应用场景与实施模式均较为有限。总体而言，针对亚太地区开展的网络欺凌干预元分析结果显示，现有方案在降低网络欺凌施害与受害行为方面未产生统计学显著性效果。