Collecting a Representative Sample of Social Movement Events to Study Change Over Time

社会运动活动是民主不可分割的一部分，让人们有机会走到一起，集体表达对民选官员和政府政策的担忧。尽管学术界对这些事件的关注由来已久，但由于方法的限制，直到最近还没有具有全国代表性的样本存在。最重要的是，报纸--先前研究的首选数据来源--不成比例地报道了规模更大、更具争议性的事件，对其社会动态产生了歪曲的看法。2010-11年度对这些事件的全国调查改变了这一点。有了有史以来第一个具有全国代表性的这些事件的样本，研究人员取得了许多新的发现，比如捕捉到了事件的平均规模，以及事件在全国范围内动员起来的最常见原因。通过收集另一波这样的事件并将其与最初的事件进行比较，该项目在研究社会变化方面开辟了新的天地。该项目将分析两种类型的变化。首先是事件特征的变化。与第一波相比，第二波研究中的事件规模是更大还是更小？与八年前相比，今天的活动是否更多地聚焦于某些原因(例如，枪支暴力、移民或妇女问题)？第二，是解释性的改变。与第一次调查相比，现在的某些因素是更强还是更弱的预测因素？例如，与八年前相比，参加活动的黑人比例今天产生的警察人数是更多还是更少？调查结果将有助于了解民主生活和民间社会，使民选官员和政府机构能够掌握有关集体行动轨迹的重要信息，从而能够更好地规划公民安全和参与。社会运动活动反映了与会者对政府政策的担忧，但由于缺乏可靠的数据，跟踪这些活动的变化一直受到阻碍。该项目建立在产生这些事件的代表性样本的早期工作的基础上；该项目将进行第二次这样的调查。第二波调查将采用超网络抽样，抽取1000多名具有全国代表性的活动参与者作为样本，回答有关他们参加活动的客观特征的问题，如出席人数、日期、地点、使用的策略、组织赞助以及警察和其他公民在场。重复的事件将被删除。此外，将制定和应用统计权重来调整：(1)在超级网络抽样的情况下，较大活动被纳入的可能性更大；(2)一些公民在过去12个月中参加了不止一次活动，因此并不是所有活动的入选概率都相同。活动将按城市进行地理编码，并与人口普查-地点级别代码相关联。然后，将人口普查变量(例如，人口规模)附加到它们，以便可以检查上下文变量。为了分析两个样本之间的变化，将使用双变量(t检验)和多变量(回归)统计模型。此外，将采用数据可视化技术，以便于解释和了解事件随时间发生的变化。这些发现将推动关于社会运动动力学的社会学理论的发展，并对更多地了解最有可能激励集体行动的社会问题类型产生影响。调查结果还将为有关民主参与的几个领域的更大文献提供信息，特别是关于事件特征和轨迹的背景差异，从而为社会变化理论提供信息。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Social movement events are an integral part of democracy, giving people the opportunity to come together collectively to voice concerns about elected officials and governmental policies. Despite longstanding scholarly attention to these events, until recently no nationally representative sample of them existed because of methodological limitations. Most significantly, newspapers--prior research's go-to data source--disproportionally cover larger and more contentious events, generating a skewed view of their social dynamics. A 2010-11 national survey of these events changed that. With the first-ever nationally representative sample of these events in hand, researchers made many new discoveries, such as capturing the average size of events and the most common causes for which events mobilized across the nation. By collecting another wave of such events and comparing it to the original one, this project breaks new ground in the study of social change. The project will analyze two types of change. First is change in the characteristics of the events. Are event sizes in the second wave study larger or smaller relative to those in the first wave? Do today's events focus around certain causes (for example, gun violence, immigration, or women's issues) more often than they did eight years ago? Second is explanatory change. Are certain factors stronger or weaker predictors now than in the first survey? For instance, does the proportion of blacks attending events generate more or less police presence today compared to eight years ago? Findings will inform understanding of democratic life and civil society, allowing elected officials and governmental agencies to have important information about trajectories for collective action, thus enabling better planning for citizen safety and participation. Social movement events reflect attendee concerns with governmental policies, but tracking changes in these events has been hampered by lack of reliable data. This project builds on earlier work that produced a representative sample of these events; this project will conduct a second such survey. The second survey wave will employ hypernetwork sampling to draw a nationally representative sample of over 1,000 event attendees to answer questions about objective features of events they attended, such as turnout, date, location, tactics used, organizational sponsorship, and police and other citizen presence. Duplicate events will be removed. Additionally, statistical weights will be developed and applied to adjust for: (1) the greater probability of inclusion for larger events given hypernetwork sampling; and (2) the fact that some citizens attend more than one event in the last 12 months, and thus not all events have an equal probability of selection. Events will be geocoded by city and linked to Census-place level codes. Then, Census variables (for example, population size) will be attached to them, so that contextual variables can be examined. To analyze change between the two samples, both bivariate (t-tests) and multivariate (regression) statistical models will be used. Moreover, data visualization techniques will be implemented to facilitate interpretation and understanding of changes in events over time. These findings will advance sociological theories about social movement dynamics, with implications for greater knowledge regarding the types of social issues most likely to motivate collective action. Findings will also inform larger literatures in several fields regarding democratic participation, particularly regarding contextual differences in event characteristics and trajectories, thus informing theories of social change.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.