This paper presents the first large-scale, multi-source analysis of the outcomes of China's extensive selective censorship effort. The authors developed a system to locate, download, and analyze the content of millions of social media posts from nearly 1,400 Chinese social media services before the government could censor them. Using modern computer-assisted text analysis methods, they compared the content of censored and non-censored posts across 85 topic areas over time. Contrary to previous assumptions, posts with negative or vitriolic criticism of the state, its leaders, or its policies were not more likely to be censored. Instead, the censorship program was aimed at preventing collective action by silencing comments that represented, reinforced, or spurred social mobilization, regardless of content. The study shows that censorship is oriented toward preventing collective activities that are occurring now or may occur in the future, revealing the government's intent. The paper discusses two theories of Chinese censorship: the state critique theory, which posits that the goal is to suppress dissent and criticism of the state, and the collective action potential theory, which suggests that the target is people who join together to express themselves collectively, potentially leading to collective action. The authors find that the collective action potential theory is correct, as censorship is primarily aimed at restricting the spread of information that may lead to collective action, regardless of whether the expression is in direct opposition to the state or related to government policies. The study's data collection involved locating, accessing, and downloading posts from 1,382 Chinese websites during the first half of 2011. The authors used stratified random sampling to select 85 topic areas based on political sensitivity. They analyzed the content of posts, identified volume bursts, and categorized events into five content areas: collective action potential, criticism of the censors, pornography, government policies, and other news. The results showed that censorship was most intense during volume bursts associated with events with collective action potential, such as protests, criticism of the censors, and pornography. Censorship was less intense for events related to government policies and news. The authors also found that the content of censored and uncensored posts was similar, with both critical and supportive posts being censored when events had collective action potential. The study concludes that the Chinese government's censorship program is aimed at preventing collective action by silencing comments that represent, reinforce, or spur social mobilization, regardless of content. The findings provide new insights into the Chinese government's intentions and the effectiveness of its censorship program.This paper presents the first large-scale, multi-source analysis of the outcomes of China's extensive selective censorship effort. The authors developed a system to locate, download, and analyze the content of millions of social media posts from nearly 1,400 Chinese social media services before the government could censor them. Using modern computer-assisted text analysis methods, they compared the content of censored and non-censored posts across 85 topic areas over time. Contrary to previous assumptions, posts with negative or vitriolic criticism of the state, its leaders, or its policies were not more likely to be censored. Instead, the censorship program was aimed at preventing collective action by silencing comments that represented, reinforced, or spurred social mobilization, regardless of content. The study shows that censorship is oriented toward preventing collective activities that are occurring now or may occur in the future, revealing the government's intent. The paper discusses two theories of Chinese censorship: the state critique theory, which posits that the goal is to suppress dissent and criticism of the state, and the collective action potential theory, which suggests that the target is people who join together to express themselves collectively, potentially leading to collective action. The authors find that the collective action potential theory is correct, as censorship is primarily aimed at restricting the spread of information that may lead to collective action, regardless of whether the expression is in direct opposition to the state or related to government policies. The study's data collection involved locating, accessing, and downloading posts from 1,382 Chinese websites during the first half of 2011. The authors used stratified random sampling to select 85 topic areas based on political sensitivity. They analyzed the content of posts, identified volume bursts, and categorized events into five content areas: collective action potential, criticism of the censors, pornography, government policies, and other news. The results showed that censorship was most intense during volume bursts associated with events with collective action potential, such as protests, criticism of the censors, and pornography. Censorship was less intense for events related to government policies and news. The authors also found that the content of censored and uncensored posts was similar, with both critical and supportive posts being censored when events had collective action potential. The study concludes that the Chinese government's censorship program is aimed at preventing collective action by silencing comments that represent, reinforce, or spur social mobilization, regardless of content. The findings provide new insights into the Chinese government's intentions and the effectiveness of its censorship program.