悠悠2017-03-02 1:31 PM

Economics / Finance


This is a lay summary for the article <ARE PROTESTS GAMES OF STRATEGIC COMPLEMENTS OR SUBSTITUTES? EXPERIMENTAL EVIDENCE FROM HONG KONG'S DEMOCRACY MOVEMENT> that was first published on <Journal of Political Economy> on February 13th, 2017. Authors of this research include Professor Y. Jane Zhang from Division of Social Science Hong Kong University of Science and Technology. 


What drives Hong Kong students’ decisions to participate in the high-stakes Hong Kong's pro-democracy movement?

At the core of social science research on protests lies a crucial strategic element: an individual’s participation is shaped by beliefs about the participation of others. Much recent work assumes that the protest game is one of strategic complements: beliefs that others are more likely to turnout increase one’s own likelihood of turnout. Yet, incentives to free-ride on others’ political collective action or to send a signal of one’s “type” or “identity” could generate a game of strategic substitutes.

While much theoretical work has been done on the strategic element of the protest decision, empirical evidence on the causal effect of beliefs regarding others’ protest turnout on one’s own is extremely limited. This paper conduct the first experiment studying individuals’ strategic motives to participate in a high-stakes political protest, manipulating the beliefs of potential protest participants.


The goal of the experimental design of this research is to isolate the causal effect of variation in beliefs regarding others’ protest participation on one’s own protest participation. To do so, we provide a random subset of individuals in our sample truthful information intended to shift beliefs regarding others’ protest participation. A challenge we face is that such information must be provided prior to the protest itself — before we know the actual protest decisions of others.

To solve this problem, we issues three part studies as follows:

-One week before the protest, on June 24, 2016, researchers collected information on individuals’ own planned participation in the upcoming July 1 anti-authoritarian protest, individuals’ beliefs about others’ planned turnout, as well as individual ’ beliefs about other’s future actual turnout at the protest. In this part study, researchers recruited participants sending an email to the entire undergraduate population of the Hong Kong University of Science and Technology (HKUST). Researchers received 1,744 completed surveys, Among these, they focus on the 1,576 students who were either born in Hong Kong or moved there prior to high school (Hong Kong “natives”). they paid students for their participation, On average, respondents received HKD 205, approximately US$ 25, for completing this first survey.

-On June 30, 2016, researchers provided a random subset of their experimental sample with information regarding the true level of planned protest participation (information obtained from the first survey). For both the information treatment group and the control group, researchers again elicited beliefs regarding other subjects’ actual protest participation on July 1, 2016. Comparing posteriors between the treatment and control groups provides an estimate of the “first stage” relationship. Experimental intervention was also conducted in Part 2 of the study, a very short online survey sent in an email on June 30, 2016, and completed by 1,303 Hong Kong native students. students received a payment of HKD 25 for completing the survey.

-On July 15, 2016, Researchers elicited subjects’ participation in the July 1 protest. This provides them with their outcome variable of interest and comparing participation rates between the treatment and control groups provides an estimate of the “reduced form” relationship of interest. Self-reported July 1 protest participation is also the outcome in their two stage estimates of the effects of beliefs regarding others’ protest participation on one’s own. Researchers elicited students’ participation by sending a third online survey via email on July 15, 2016, and completed by 1,241 Hong Kong native students.19 Students who completed Part 3 of the study received an additional payment of HKD 25.


The effect of the information treatment — informing the treatment group that 17% of experi- mental subjects planned to attend the protest — can be seen in the distributions of beliefs regarding subjects’ actual participation, presented in Figure 2. One can see in the figure that prior beliefs regarding actual participation for the majority in the experimental sample were below 15%; the median is at 10%, and the distribution exhibits a long tail. One also sees in the figure that the distributions of posterior beliefs regarding actual participation look very different comparing the treatment and control groups. The control group’s distribution of posteriors looks very much like the experimental sample’s distribution of priors, but shifted slightly to the right. In contrast, the treatment group’s posteriors are distributed much more tightly between 10% and 20%.

Researchers find a broad range of evidence indicating that Hong Kong students considering participating in the July 1 protest of 2016 viewed the strategic element of their decision as a game of strategic substitutes. Experimental exposure to information regarding other subjects’ planned participation affects Hong Kong university students’ beliefs regarding other students’ actual future participation in an antiauthoritarian protest, and affects students’ own protest participation. Specifically, individuals in the treatment group with prior beliefs below the truthful information provided to them updated their beliefs positively and became less likely to participate in the protest; individuals with prior beliefs above the truthful information provided to them updated their beliefs negatively, and became more likely to participate. The negative association between beliefs regarding others’ participation and one’s own is also seen in the naturally occurring variation in beliefs present among control subjects in this study.


Davide Cantoni, David Y. Yang, Noam Yuchtman, Y. Jane Zhang. (2017, February 13). Are Protests Games of Strategic Complements or Substitutes? Experimental Evidence from Hong Kong's Democracy Movement. National Bureau of Economic Research, 10.3386/w23110.






主编 唧唧堂 内容运营

唧唧堂主编,热爱八卦和运动,学术的话最近最感兴趣的是各种医学的研究~~欢迎大家加我个人微信:hiyouyouxuan, 一起探讨如何把唧唧堂变得更好~~

11 Following 23 Fans 0 Projects 56 Articles


几十年的互联网发展已经证明其是人们信息共享强有力渠道,并且开始成为实现所谓的“群众的智慧”的重要根据地。互联网经济新近的一个重要的趋势是,各种评价系统不断涌现,这种评价系统有针对餐馆酒店的,比如大众点评; 有评价旅游景点的,比如穷游网; 有打算评价教授大学的,比如唧唧堂,甚至有评价宗教人士的,小编暂时

Read More

我们都知道,即使在产业内,公司的生产率也是不同的。 在国家层面这种差别更大,特别是当我们比较处于不同经济发展阶段的国家时。另外,在资本的边际收益产品中我们也可以发现这种生产率的差别。 该论文发表之前已经有研究试图从不同行业国家经济变量参数的边际产品波动中识别出资源的不当配置程度,以及这种不当配置对社会

Read More

没有一个城市是单一产业的城市,就连好莱坞或者硅谷都不是。 同样,也没有任何一个城市是拥有完全不同产业的。 传统的城市系统模型聚焦的城市非常两极化,要么研究产业完全专业化的城市(completely specialized city),要么研究完全不同的城市(completely diverse cit

Read More

人类可以通过查看一个概念(concept)的一个或者几个简单的例子(instances)就能够识别出更多其他例子,更具体说,例如人第一次看到一头猪,就能识别这头猪的一些概念特征,耳朵很大,鼻子突出,大鼻孔等等,当人得知长这样的物体是猪之后,下次他去农场,就能将猪认出来,就算那里还有更多其他家畜。 这是

Read More


Read More

每年12月,英国伦敦的Digital Science公司都会评选出该年最热门的100篇论文,他们的评论标准,并不是根据论文发表杂志的影响因子,而是利用他们自己建立的Altmetric评价系统, 基于论文在媒体、博客、推特、脸书和维基百科等类非学术共同体被引用或转发的情况而定的,所以选出的论文不仅有重大

Read More

Moodle 设计开发是依据以下对 “社会构建教育”的理解展开的。 对社会教育的理解包括以下几个方面,我觉得对唧唧堂的发展应该也有知道意义。 Constructivism (构建主义) From a constructivist point of view, people actively

Read More

悠悠 Sep 06 | 06:40 PM


创造产业是许多科技人员的梦想,虽然这很不容易做到。一位很优秀的博士生,导师是“大牛”,在国际学术界有一定知名度。博士生已在IEEE Trans发表了多篇论文,有的还在国际会议上得奖。但面临毕业后的选择,仍然很艰难。因为他清楚,继续堆SCI可以在科技界呆下去,但处在应用科学界,原创性贡献的方向在哪里?与

Read More

This is a lay summary for the article <Repeated Interactions and Improved Outcomes: An Empirical Analysis of Movie Production in the United States> tha

Read More


Read More