btw, imy!

︎︎︎ 12 november 2020
The purpose of this excerise was to create data visualizations formatted for Instagram. Then, the post had to be refuted.

"btw, imy!" argues that students send more texts when they can't see their friends whether online or in-person. However, the claims are based on speculation and don't have enough data points to prove the validity of the statement.

To begin, the first data visualization claims that "text messages received reached their peak (300-599) during the middle of the week (Tuesdays and Wednesdays)." However, there were 2 Wednesdays (of the 4 Wednesdays recorded) that only hit the 100-200 range. Depending on how you look at it, the bar chart supports the stated claim 50% of the time while refuting it the other 50%. This assertation that "text messages received tended to decline during the weekend" continues into the third data set.

The fourth data set maps out food purchases in Fayetteville across the four week timeline. It states that "food and drink purchases are spread out around sociable places in town, like MLK and Dickson." The word "sociable" can be called into question here since it’s not explicitly defined. Based on the data received (but not published on these slides), only some of these purchases were made at sit-down establishments on MLK. The majority of MLK purchases were made at fast-food places—establishments that people traditionally drive-thru or carry-out. Additionally, a portion of these transactions from businesses on MLK were made via Door Dash. So it's likely that over half of the meals or drinks from MLK weren't actually shared with friends. Another weak point of this map is that it's limited to the Fayetteville area. There were a handful of purchases made in Bentonville and Texas that had to be omitted, warranting an incomplete story.

The claims made in the data visualization may only be able to provide a general idea of social behavior but can't be used to make definitive conclusions without further investigation. There are a handful of days that don't support the data set as shown. For example, take Wednesday the 9th on the calander. Text messages received were high (501) and Instagram was also the most used app (additionally, food and drink were purchased). The main point that this data set is trying to argue is that either social media apps are used a lot or texts are used a lot—it can’t be both on the same day. So, to support the overarching claim, we'd have to know some along the lines of: even though Instagram was the most used app, it was only used for 20 minutes that day. We're unable to understand relationships with the exclusion of quantitative data on app usage. To affirm "btw, imy!," it would be helpful to know the content of the messages (both texts and social media DMs) as well as exactly how many people this person saw throughout the week.

The platform this was formatted for—Instagram—limits the amount of information you can squeeze into 10 1080 x 1080 px slides. A platform that allows for more robust details should be used to illustrate the statements made in "btw, imy!" Perhaps the one thing that undermines these slides the most is the fact that only one person's experience was used to create these data sets. To ensure validity, more people would have to record their food transactions, text messages, and app usage in greater detail.