The introduction of groups and you may good pilot accuracy attempt endured 3 days, during which date the new codebook are modified once or twice. So you can shot accuracy, several programmers (scholar students) alone coded a group from 100 randomly chosen meme circumstances (on ten% of one’s decide to try). The fresh new agreement rates is actually looked at using the Krippendorff ( 2004b ) decide to try, with .66 given that minimum get to just accept an adjustable because necessary (Krippendorff, 2004a ). Eventually, a couple of variables, subculture (content) and you may photo number (form) obtained an inadequate get, and was indeed taken from the latest codebook, leaving forty eight parameters into genuine coding (Discover but in addition for the entire codebook of simple functions when you look at the our data). After reliability is established, the rest of the attempt are split up between them programmers.

System investigation

I working community study processes (Barabasi & Frangos, 2002 ; Newman, Barabasi, & Watts, 2006 ) to help you map the content, participation, and function similarities between meme occasions as well as their cohesiveness. Our very own stars regarding community had been the fresh meme instances, and their undirected links was indeed the common content, participation, and site de rencontre pour agriculteurs aux usa you will setting qualities. Such as, when the meme times A beneficial and you may B display around three simple qualities in preferred, its tie manage lbs three. I created three independent undirected channels to have articles, means, and contribution, in addition to a system that combines every issue with her. The latter stands for the overall distance certainly one of meme occasions in our attempt.

Network analysis enabled distinguishing the essential central meme period, to put it differently, people who shared the greatest quantity of blogs, involvement, and you can means functions with individuals, as well as individuals who was in fact so much more peripheral, and thus got unique or rare functions ( ). While doing so, looking at the general build of your community, we could select clusters out of meme times with a high cohesiveness from content, participation, and you may function. We could up coming evaluate hence meme family members have been far more cohesive than simply someone else and you will show the web link between general qualities and you may nearest and dearest cohesiveness.

Measuring the fresh new cohesiveness off meme group

Structural cohesion inside the networking sites enjoys some dimensions (White & Harary, 2001 ). I chose the most simple you to, and you may compared brand new connections one of instances of the same family with connections outside the family unit members. The newest cohesiveness changeable was created to mention to what the total amount meme era were just like other occasions regarding the exact same family from inside the terms of simple posts, involvement, and you can form. Earliest, i computed the fresh system training (Barabasi & Frangos, 2002 ; Newman, Barabasi, & Watts, 2006 ) each and every meme particularly, i.e., the number of popular content, participation, and you can form characteristics this distributed to other days in the exact same friends sufficient reason for period from other household. Following, for every of the fifty meme group in our sample, we determined the fresh imply amount of preferred attributes the circumstances shared within this and beyond your family unit members. In the end, the fresh cohesiveness variable is actually recognized as the difference regarding imply amount of preferred attributes in this and you may outside a good meme friends.

Computing the newest individuality out of meme household

The new dimension from individuality level was designed to speak about how book a great meme household members was a student in regards to the simple articles, involvement, and you can form properties. To put it differently, uniqueness top mirrored the new pit between your common features of a good meme family members and traditional (the most used universal functions about entire decide to try). Very first, i calculated the latest part of stuff, participation, and you can form features within the per family; next, i computed this new part of for every trait on entire decide to try; finally, per meme family, i calculated the newest square root of your own sum of this new rectangular differences between the 2 (as in the algorithm to have simple departure). It was standard so you’re able to include 0 and step one, where step one signifies the number one individuality out-of a good meme nearest and dearest during the terms of its general attributes.