The unexpected findings of homophily and fashion

Homophily and the Clothing industry seem to have a pretty simple relationship, right? We spend our time with like minded individuals many of which share our tastes in dress and that is what stimulates this industry, right? According to Cao,Qin,Yang, and Zhang’s findings in their journal titled “Fashion and Homophily”, surprisingly a saturation of Homophily is what hinders this industry and Fashion cycles (the driving force behind the industry at large) is product of Heterophily.

In this article they explain the importance of planned obsolescence on driving an economy. The problem with Fashion (which happens to be a huge contributor to economic growth and stability) is that most people (with the exception of the very wealthy) can only justify an expensive purchase if it means that the item they receive in return is going to last. Since the products in the industry are designed to last, designed obsolescence would not be an effective strategy to drive the industry forward. Instead the fashion industry adopted clever techniques to make their non-disposable products “last seasons news”.

Think about it like this: If everyone dressed the same and companies always made the same products then nobody would need to buy another piece of clothing until the stuff they have is ruined or until they don’t fit them anymore. Homophily is detrimental to the growth of the fashion industry because it allows for people to remain socially comfortable wearing the same clothes as everyone else until they cannot be worn any longer. This is why Heterophily is important for this market, it allows diversity and it allows the “rebels” (as they call them in the article) to pave the way for new consumer tastes and trends and fuels the innovation and momentum the industry needs to survive and thrive.



Does Homophily Play a Role in the Formation of Cliques?

A prime example of homophily are the cliques that form during high school. These cliques consist of individuals who are very similar in their personalities and in many ways share the same interests. In the article “Why Cliques Form at Some High Schools and Not Others” by Derek Thompson, the reasoning behind why this occurs on the high school level is discussed. The interesting points that this article explains are some of the factors that cause cliques to occur. Derek points out that the main reason cliques form is due to the environment of the high school itself. If the high school has an environment that actively has the students interacting with one another it is very unlikely that cliques will form. Since, this is not how most high school environments are students have no reason to move out of the comfort zone that homophily creates for them. The article points out that it is most likely for cliques to develop in large schools. This is because groups of individuals can exist separately from one another, yet have no overall affect on the school. Additionally, over time cliques under these circumstances will become more pronounced and segregated. This result is directly demonstrated by the Schelling Model, in that over time groups of individuals will naturally segregate themselves if conditions allow. The occurrence of cliques on a high school level is a clear and pronounced representation of homophily, and helps give insight into why segregation occurs. The last thing of note is why cliques really only occur in high school and not in other environments. The main reason for this is that segregation and group formation is not discouraged through forced social interaction. In many other environments individuals are “forced” to interact with other individuals that may not be in their “group”. This limits the degree to which cliques can form. Given this information it can be seen that cliques are actually a natural part of social networks due to homophily, not just a strange phenomenon that occurs in high school.

Article link:

You Might Know . . . (Homophily)

Below is an interesting article on homophily published by the New York Times in 2006. In essence, the Times looks at the many topics homophily is being used to interrogate (dating, friendship circles, buying preferences) and how social networks were using homophily on their sites. The article concludes by questioning if homophily on the Internet was positive and offers an alternative site which aims to put people who do not have much in common together.

Nine years later, we still see social networks using homophily; especially Twitter and Facebook with their “You Should Follow” and “You May Also Know” features. These features not only connect you with people you know, but also make it more likely for you to continue to use the social network as it will consistently be updated with information that is relevant to your real life. While using social media to meet people with different interests sounds interesting, constantly having updates about people you don’t know or have little to nothing in common with does not seem like it will keep your interest. Moreover, dating someone who has little in common with you also seems unappealing. In many ways, it’s homophily that has allowed social networks to work. They have found ways to connect you with people you already know or those with similar interests.

Social Media is Making Us Dumber?

Time magazine published an article titled ‘Social Media is Making You Stupid’ (February 21, 2014) based on findings from a new study published in the Journal of the Royal Society Interface. The study was concerned with the transfer of information through networks, particularly social media networks, and the effects it may have on the individuals within them. This study concluded that highly connected networks share information the most efficiently (due to the number of links between members of those networks) through a study where a group of 100 people were split into 20 person social networks connected in various patterns (one extreme where all were connected, the other where all members were isolated) and asked to answer a series of questions. The members of the networks were asked the questions five times; the first time they had to answer the questions on their own, but for the rest they were able to copy their neighbor’s answer. This experiment was used to show how the more highly connected one is to other sources of information, the more easily available that information is (the more highly connected social networks were better at answering the questions the more times they were asked/able to copy their neighbor’s answers), but it also shows that gaining information in this way does not necessarily display understanding of the information that is gained through highly connected social networks.

The article claims that social media has a similar effect; pieces of information are constantly being passed through social media, members of those networks are gaining information, but they are not necessarily gaining the analytic skills to discover that information on their own. Thus, using social media “may very well decrease the frequency of analytical reasoning by making it easy and commonplace for people to reach analytical response without engaging analytical processing”.

This article appears to address the concepts of selection and social context/socialization that we discussed in class by looking at some of the implications of that transfer of information. In the experiment, the way that one could copy the answer of their neighbor when answering the questions is an example of a kind of socialization where one adopts the behavior (answer) of their neighbor, while the choice of copying one neighbor’s answer over another is an example of selection. But the experiment discussed raises interesting questions about the kind of superficial quality that information dispersed through networks can assume. In the case of information shared on social media sites (whether it be political, economic, social, etc.), networks of often highly connected individuals, members are exposed to a huge amount of information that may be retained in a similar way to the copying of a neighbor’s answer in the experiment. So, while information may travel far and wide throughout a well-connected network, the processes leading to that information may be left untouched; leading to a lack of understanding of those processes. This raises interesting questions concerning selection and socialization, as individuals within a highly informed and highly connected network (social media) may appear to be getting smarter due to the availability of information, but in reality they may only appear to be getting smarter as they are simply undergoing the process of selecting options from their highly informed social context; the information is made available to members of social networks but they may, in this exchange of information, not actually gain an understanding of that information. Members of highly informed and connected networks are able to absorb a large number of facts, without learning the analytic processes leading to those ‘facts’; a superficial knowledge of facts appears to be one of the possible results of these kind of highly connected and highly informed networks. Interesting.


Benjamin Stieler





Homophily and Police Diversity

An important lesson in homophily and minority representation in law enforcement follows the recent death of teenager Michael Brown in Ferguson, MO.  According to, the town, despite being 60% black, has only three black police officers, and there is evidence to suggest that a police force that is not reflective of its community creates more tension and brutality than a more representative force.

Nationally, whites are overrepresented in law enforcement by 30 percentage points.  It is clear that a police force that looks like the community it protects leads to less police violence and a greater feeling of overall safety in the community.  Los Angeles is given as an example in the article because the force has, in recent years, undergone serious change to diversify.  According to a Kennedy School study, “We found the LAPD much changed from eight years ago, and even more so in the last four or five years. Public satisfaction is up, with 83 percent of residents saying the LAPD is doing a good or excellent job; the frequency of the use of serious force has fallen each year since 2004.”

A main reason that more diverse forces are more effective is the level of trust of the community.  Community members exhibit homophily, meaning that they feel safer and more connected to people who look and act like they do.  While black officers may be able to better communicate with and get cooperation from black citizens, it is clear that most of the benefit of the diverse force comes simply from the fact that the law enforcement looks like the community it is representing and therefore exhibits homophily.  According to a sergeant in Atlanta, “…so every race, gender, sexual preference, religion, whatever it is, you can look up in the command staff and identify someone who is the same as you. And if you need to reach out to anyone with a problem, someone in that command staff is similar to you with whom you can have a candid conversation.”


Homophily Won’t Let Us “Let It Go”

If there is one song that I have to say was the hit song of 2014, it would be “Let It Go” from the movie “Frozen.” I myself have yet to see the movie, which premiered in 2013, even though my family and the rest of the world, it seems, already have numerous times. Yet, this song, that one line “let it go, let it go,” is embedded in my head forever; the movie, the characters, the song— they are everywhere! So why can’t we escape the Disney’s phenomenon “Frozen”?

Psychologists Yalda Uhls and Maryam Kia-Keating, who are sisters by the way, explain to CNN why the movie continues to be so popular. I learned that the two main characters are sisters named Anna and Elsa. Uhls states that the family-themed message in the movie resonates with little children. Homophily, the principle that we tend to be similar to our friends, is apparent here with children finding connection to the ties of the two sisters; kids can easily identify with Anna and Elsa’s friendship. Kia-Keating adds that because of this, preschoolers have a “loyalty and unrelenting interest to watch this movie over and over again.” Elsa also has powers, and according to Uhls, children feel empowered when a character they can relate to has powers, since said children are always told what to do. Preschoolers’ behaviors are also much like Anna’s and Elsa’s, impulsive with lack of emotional control. Finally, the message behind the “Frozen” theme song, “Let It Go,” is about being a good girl or boy, a message kids often hear so they copy it.

Next time you see Anna and Elsa anywhere or hear my hit song of 2014, remember it’s only the effect of homophily at its best (especially for Disney with “Frozen” being its highest-grossing animated film).

Divorce Contagion

“Is Divorce Contagious” by Rich Morin highlights the triadic closure property, contagion within a network, selection, and socialization in divorce. The article cites a study which focuses on data gathered over 30 years from couples who married, divorced, and then remarried in Framingham, Massachusetts. The study concluded that when one couple gets divorced, there is a 75 percent chance that one of their close married friends will  divorce.  The study also found that friends of friends of the divorced couple had a 33 percent more chance of  divorcing.  This statistic  can be expressed using the triadic closure property in a network.   The couples with strong ties to the divorced couple had a 75 percent chance of divorcing while those with weak connections to the divorced couple had a 33 percent chance of divorcing.

The study is also  applicable to the economics term, contagion within a network.  Divorce, a contagion within the network, was evident when 75 percent of couples with strong ties to the original couple divorced.  Divorce is a contagion because when one couple  divorced, their close friend had a higher chance of divorce.  If that close friend got a divorce, then  that couple’s friends had a 75 percent chance of  divorcing and the trend continues.

I believe the chance of divorce increased as a couple’s friends got divorced because of selection.  After the study’s participants saw their friends  divorce, they more clearly realized their own marital problems. This realization could have possibly led to their divorce.

Socialization could have also been the cause of the friend’s divorce because people try to become more like  their friends.  If a friend divorces, it may lead all the other friends to divorce because they want to be like their divorced friend. Overall, this is a compelling article about how networks operate all around us without us even noticing.