Hurricane Harvey was social media's first real stress test. Despite headlines that praise platforms like Facebook, Twitter, and Nextdoor for their role in disasters, we still know little about how community resilience differs across these platforms in a single event across a single measure of recovery. With an original dataset from a survey administered in Houston 39 days following Hurricane Harvey, this study explores the community resilience outcomes across users of Facebook, Twitter, and Nextdoor; more specifically, if users agreed that their neighbors engaged in neighborhood-level coordinated recovery efforts in the weeks following the storm. Controlling for collective efficacy, storm impact, amount of aid received, age, gender, income, education attainment, home tenure, and survey modality, findings from two multinomial logistic regression analyses reveal that individuals who used Nextdoor were more likely to receive help from neighbors to collect and pile debris following Hurricane Harvey compared to individuals who indicated they never used the platform, a finding not found among Facebook and Twitter users. This study identifies important differences in the resilience outcomes across these three platforms and discusses implications for future research and resilience policies.

飓风哈维是社交媒体的第一个真正的压力测试。尽管头条新闻称赞 Facebook、Twitter 和 Nextdoor 等平台在灾难中的作用，但我们仍然知之甚少，在单一事件中，在单一恢复措施中，这些平台之间的社区复原力有何不同。本研究使用哈维飓风过后 39 天在休斯顿进行的一项调查的原始数据集，探讨了 Facebook、Twitter 和 Nextdoor 用户的社区复原力结果；更具体地说，如果用户同意他们的邻居在风暴后的几周内参与社区级别的协调恢复工作。控制集体效能、风暴影响、收到的援助金额、年龄、性别、收入、教育程度、家庭保有权和调查方式，两项多项逻辑回归分析的结果表明，与表示从未使用过该平台的个人相比，使用 Nextdoor 的个人在飓风哈维之后更有可能从邻居那里获得帮助来收集和堆放碎片，这一发现在 Facebook 和 Twitter 用户中没有发现。本研究确定了这三个平台的复原力结果的重要差异，并讨论了对未来研究和复原力政策的影响。