How can we use the mechanics of influence to drive behavior for public good?
When one thinks about social influence, he or she often imagines well-known people, such as celebrities, politicians, and athletes. While it is absolutely correct that they have major impact on various parts of the world, our everyday lives are usually affected by people who we know well and are close to us, such as family, co-workers, and fellow students. For example, when we try to buy clothes, our decisions are affected by clothes that our friends wear, so that we are not perceived in certain undesirable ways.
Social influence is being measured in various ways by different social media networks. Yahoo, for example, estimates a “social authority score” for each user, calculating how large the user’s network and his or her friends’ networks are. Facebook measures how fast a piece of content is shared within the personal network of a single user, so that it can find out which users are capable of exerting an extent of influence among their peers.
However, the traditional models of social influence have two main flaws: causality of behavioral change and homophily. The first flaw, causality of behavioral change, can be seen in many examples in today’s world. Celebrities, for instance, may have many followers on Facebook and other social media, but those followers usually never do what the celebrities advise them to do. If a celebrity has many “followers”, but he or she doesn’t affect their behavior, how is that celebrity exerting real social influence? Homophily is defined as an individual’s tendency to be associated with people that have much in common with that same individual. Some followers of a particular person may do as that individual advises, but not necessarily because that individual has an influence over those people; the people are doing what they are doing because they already have a similar mindset as that person. In short, for a particular individual to truly have social influence on others, he or she must directly affect how his or her followers act.
Thus, these are some proposed solutions to improve the mechanisms of social influence of businesses and other policy makers for the benefit of the public. First, it would be effective to design selective exposure to certain groups with desired behaviors. People who engage in a particular type of behavior are much more likely to respond to a relevant exposure. Second, selective exposure is best designed at early opinions to cause “herding” effects. For example, MIT researchers discovered that a single like of an online story resulted in a 25% increase in overall ratings. Lastly, incentives should be designed to strengthen peer-to-peer influence. In the DARPA Network Challenge in 2009, competing teams of scientists were assigned to find 10 red weather balloons in different areas around the US. To gather volunteers, the winning team from MIT forwarded the message that they will promise to share the $40,000 prize equally among the team members, and the message spread and spread.
This article gives many examples of strong and weak network connections, which can be applied to the topic of network structure that was covered in this course. Homophily is a factor of weak links, because a follower that is connected to a certain individual is not being directly influenced by that same individual; they are like-minded in their actions. Also, we can also learn from this article that strong ties arise mostly from people whom we interact with every day and are around us, such as friends and family. Structural balance of networks can be attributed to the recommended solution of selective exposure to certain groups with desired behaviors. The edges of all of the triangle networks in these certain groups are likely to be positive, because everyone here has a common type of behavior, thus satisfying the property of network structural balance.
Source: