Information Cascades and Public Opinion
Recently in class the idea of information cascades has been broached on multiple occasions. Information cascades exist as readily in real life as they do on paper, and it is important to recognize how easily this phenomena can infiltrate daily life because sometimes the wisdom of the crowds can detract from and reduce individual judgment, regardless of whether or not the decision being made by the vast majority is actually “right.” One example where information cascades often appear is in politics, where they may be referred to as “opinion” cascades, perhaps based on the subjective nature of the questions typically being asked. Neil Malhotra, a professor at Stanford graduate school of business, and David Rothschild, an economist at Microsoft Research, explored this issue in a Washington Post article titled, “Opinion polls can be self-fulfilling prophecies.” In the article they go into further detail about opinion/information cascades and how readily they can shape public opinion.
Opinion cascades occur when there is a shift in public opinion. There are a myriad of factors which may account for this altered way of thinking, one of these being the effect of other citizens’ beliefs. This type of influence can easily be characterized as an information cascade of sorts, because it contains all the ingredients of an information cascade. First of all, people must decide their opinion on some matter at hand, perhaps for voting reasons or simply to take a stance in a day-to-day conversation. Additionally, this decision can be seen as being sequential. In today’s media-saturated society, people rarely have time to make a snap judgment before being exposed to other people’s opinions via social media sites like Facebook and Twitter. As a result, people tend to form opinions over time, after they have already acquired gradual information from others. Another factor which goes into information cascades is the idea that each person probably has some kind of private information based on their personal beliefs or information they may have acquired earlier that others do not necessarily know about. Finally, although people may learn about other people’s beliefs, they may not know exactly why the person feels the way they do, especially if the person fails to go into extensive detail, or if they fail to reveal their private information. These are all factors which make up the idea of an information cascade.
Going off of this idea, Malholtra and Rothschild conducted an experiment in which they polled people on their opinions on 3 different issues: withdrawing troops, free trade, and public financing of elections. Some of the people in the poll received “aggregated polling information showing support for the public policies ranging from 20 percent to 80 percent” (Malholtra & Rothschild). After the experiment concluded and the data was analyzed, their findings were found to support the idea of the influential nature of information cascades, because as they shifted the support for the policies, so too did people shift their opinions. There was a greater chance that people would be in favor of the opinion at hand if they knew that the general public was also in favor of this opinion.
Despite the researchers findings, there were also some discrepancies; for example, the effect of the general public’s opinion was more influential at times depending on the issue, specifically the issue of free trade. This brings to light the question of how influential information cascades are, especially during elections, when people are already divided across party lines. Is it possible that the influence of information cascades may actually be capable of overriding a person’s political party affiliation? This may be attributed to the way in which an information cascade gains momentum in the first place, because there is always the probability that the opinion being conveyed across the information cascade is not actually right or well thought out, and that people simply jumped on the bandwagon rather than considering the actual policy being proposed. This can be detrimental, because public policies should be founded on their quality rather than the opinions of others—opinions which may in fact be based on very little genuine information. Given the influential nature of information cascades, it is very important that they be recognized and studied deeper to gain a further understanding of their role in public policy and elections today.
Sources:
http://rap.sagepub.com/content/1/2/2053168014547667.full