By Eric Buehler

This post has an interactive graphic created by Bobby Craig! Click on the image below and you get to see how many swipes happened when and where across a week (by moving the sliders at the bottom of the image). If you see anything interesting, let everyone know in the comments. Then come back for more words and figures when you’re done (scroll below this image).

At the beginning of every fall semester, hundreds of Freshmen are thought to march to the Sunset Apartments in an almost ritualistic fashion to experience for themselves the party culture at Denison. On their way, they will be swept up and groups spread asunder by the various events and transactions of the night. A variety of experiences will be had fun and regrettable, good and bad, and partly dictated by one’s race and social network. While 25% of students report not being regular attendees of events at the Sunnies, the effects of party culture are nonetheless felt around Denison, especially when it comes to the mass movements of people across campus. While the 127 crew has written about the party culture of Denison, such as individual’s level of comfort at Sunnies’ party and the Greek dominated six-man’ suites of East, we have never before been able to quantify migratory patterns of campus life.

However, thanks to a new set of data, which captures a week of swipe transactions from last spring, we are able to ask questions such as what time does this migration begin? When do people return? And where are they denied entry? We do not have names and are not interested in tracking individuals, merely the aggregate movements of the student body across a week.

While visualizing these transactions is interesting in itself, it is important to take any conclusions with a grain of salt. The data used are limited to only one week, and exclude any confounding variables that may explain any given spike in the data.

Folklore tells us that we should expect to see change in swipe activities on Wednesdays, Fridays, and Saturdays. However, the week of swipe transactions organized by day (April 3-April 9 2017) shown in Figure 1 tell a somewhat different story. For example, on Wednesday, both types of Freshmen housing (quiet/special interest and regular) appear unaffected by the small spike in activity at the Sunnies and with quiet/special interest housing swipes increasing. Activity at the Sunnies on Wednesday may be largely limited to upper-class students, who most likely have friends in the Sunnies. Friday also appears to defy expectations. While the Sunnies do experience a sizable influx of valid swipes on Friday, activity at Freshmen dorms seems relatively unaffected.

It isn’t until Saturday that we see a spike in Sunnies’ activity that seems linked to activity in Freshmen dorms, with regular housing dropping from around 1000 valid swipes to a little over 800; a somewhat proportional drop is also seen in quiet/special housing. It would appear that the common notion that Freshmen flood the Sunnies on any given party night appears to be a campus legend with only one night a week linked to the decrease of valid swipes into Freshmen housing.

Moving from the days to hours, the data can be leveraged to show what a day’s worth of transactions looks like. Figure 2 reveals that while Freshmen dorms activity tapers down around 9pm, Sunnies activity simply keeps increasing, reaching an average of 35 swipes per hour by 11 pm. It appears that by 11 pm the majority of Freshmen have most likely either settled down for the night, or charted a path to the Sunnies. Additionally, it is worth noting that the Sunnies’ ability to surpass Shorney/Smith housing in number of transactions is impressive considering that the Sunnies only house 216 seniors, while Shorney/Smith house 371 freshmen.

Lastly, we can also look at when swipes are being denied. It appears that invalid swipes only become a factor after midnight when card readers start rejecting non-resident swipes at 12 am. This in itself is not entirely interesting. However, It does beg the question of how denied swipes fit into the discussion of the freshmen pilgrimage.

Figure 3 takes a close up look at the denied Swipes shown in Figure 2 by looking at the number of invalid swipes for a week divided by buildings. The figure shows that most people are being denied access to either Sunny D or Shorney/Smith, caught up in the migratory interactions of these spaces. Once again, we can note the transactions at the Sunnies eclipse those of the more populated Shorney/Smith. However, this time a single Sunny, Sunny D (Good Hall), eclipses the transactions of Shorney/Smith. To some this may show evidence that a single Sunny (D) dominates the party scene of the school. It is likely that Sunny C (Myers Hall) would most likely play this role in a more recent sample of the swipe data. Lastly, It appears that other Freshmen dorms have a much smaller number of invalid swipes, implying these dorms are outside the migratory path.

Overall, it seems that the Freshmen migration might not be as robust as lore has foretold. Nonetheless, there is still some observable evidence of the phenomena. The data suggest that this movement reaches its apex at around 11 pm, primarily occurring between Shorney/Smith and a single Sunny. By 12 am card readers begin rejecting cards, and some students find themselves caught in between the peculiar interactions between these two places.

Eric Buehler is one of those Data Analytics students who hopes to be employed one day. In the meantime, you can find him writing about student behavior at Denison.